Sign in with
Sign up | Sign in
Your question

A couple random ideas

Last response: in Video Games
Share
Anonymous
June 27, 2005 9:07:36 PM

Archived from groups: rec.games.roguelike.development (More info?)

I was bored senseless at work the other day, and started jotting down
random ideas in a notebook to pass the time. Some were terrible, some
were not overly original, but a few struck me as being interesting
enough to develop further. Granted, it'll be some time before I
actually even consider using these ideas in a game of my own, since
I'm _still_ working on my scriptable dungeon generation demo, which
should have been done weeks ago. But, anyways, I'd thought I'd toss a
few of my ideas out.

1) I don't know how many times I've found myself playing a warrior
class character, and said to myself, "Man, I wish I could get some
great ammo made, but I can't since I'm not an archer. And I really
wish I could reforge and beef up my weapons and armor, but I'm not a
weaponsmith, so I can't do that either." Various classes in games
like ADoM and ToME have great item creation/modification abilities,
but those classes are rarely the best-suited to exploit those
abilities. I always end up wishing I could give my new barbarian some
of the Arrows of Demon Slaying that my archer had created. But, alas,
I can't. Well, why not? This is the thought that formed the basis
for my idea.

Picture this: You start a new game, and get prompted to choose four
race/class combos. Next, the game world gets generated. Then, you
choose which of the four characters you want to play as right now.
All the characters share the same game world, and are affected by the
same world events. If the mighty Demon of Foo gets killed by one
character, the Demon of Foo will be dead to all characters in that
game. At any point in time, the player can save and quit, and play as
a different character in that game. The player can also exchange
items between characters in that particular game. For example, the
player could choose an Alchemist, a Blacksmith, an Archer, and a
Warrior for his four characters. He can play for a while as all of
these characters, and, when he feels ready, can prepare his Warrior to
go for the win, by equipping him with a powerful sword and armor
crafted by his Blacksmith, enchanted arrows fletched by his Archer,
and powerful potions brewed by the Alchemist. How cool would that be?
Basically, it'd be up to the player to choose which four specialized
classes he wants to use, and then use those specialized abilities to
prepare one character to go for the win.

Granted, for this idea to be fun, all the classes would have to be fun
to play as, and the item creation/modification would have to be fun as
well, otherwise the game would boil down to a tedious process of scumming.


2) In every RL out there, the player can count on seeing countless
items strewn across the various dungeons. Just lying there. In
populated dungeons. So, my question is, why haven't any of those
items been picked up by the various dungeon denizens? If I'm walking
around my apartment complex, I can count on the fact that no item
found lying in the hallways will stay there for long, because someone
will inevitably throw it away, or keep it for themselves. Now, random
items lying about in deserted dungeons makes sense. But these
dungeons are anything but deserted, and are regularly populated by
(semi-) intelligent creatures, who would likely find a pile of gold,
or a glowing dagger, or a magical potion to be definitely worth
picking up and keeping.

So, logically, any dungeon that is populated by intelligent creatures
should have pretty empty hallways. In an enemy fortress, the player
could probably count on finding weapons and armor all piled in the
armory, with some extra weapons and such lying about the barracks.
Food would probably be in a pantry, gold in a treasury, potions and
herbs in an apothecary or alchemy lab, books and scrolls in a library,
and so on. This kind of system could lead to some interesting
gameplay, I think.


3) I want monumental battles. I want to play my small part in some
epic battle at the gates of Mordor, or defending the people of Rohan
in Helm's Deep. I want to fight alonside dozens of fellow soldiers,
against a sea of enemies. I want seige warfare, catapults flinging
flaming debris and giant boulders, battering rams pounding on the
gate, soldiers digging a tunnel beneath a section of castle wall in
order to collapse it, and mighty heroes and villains clashing in the
midst of a great battle. I want all of it. But, it probably won't
happen. Such a game would need to make large-scale combat as easy to
manage as small-scale one-on-one fighting in the dungeons. It's
doable, but quite the challenge. I'd love to see a game do this
though. Nothing would be better than being able to take my level 43
Ranger and have him shoot down orc after orc from the top of the keep,
alongside dozens of fellow archers.


Well, I'm done rambling on now. This is the stuff I was pondering
over while bored, and I may actually use some of the ideas one day.
That is, once my damn walkaround dungeon generation demo gets done.


--
My projects are currently on hold, but I do have
some junk at the site below.

http://www.freewebs.com/timsrl/index.htm

--

More about : couple random ideas

June 28, 2005 1:36:58 AM

Archived from groups: rec.games.roguelike.development (More info?)

Shedletsky wrote:
> > 3) I want monumental battles. I want to play my small part in some
> > epic battle at the gates of Mordor, or defending the people of Rohan in
> > Helm's Deep. I want to fight alonside dozens of fellow soldiers, against
> > a sea of enemies. I want seige warfare, catapults flinging flaming debris
> > and giant boulders, battering rams pounding on the gate, soldiers digging
> > a tunnel beneath a section of castle wall in order to collapse it, and
> > mighty heroes and villains clashing in the midst of a great battle. I
> > want all of it. But, it probably won't happen. Such a game would need to
> > make large-scale combat as easy to manage as small-scale one-on-one
> > fighting in the dungeons. It's doable, but quite the challenge. I'd love
> > to see a game do this though. Nothing would be better than being able to
> > take my level 43 Ranger and have him shoot down orc after orc from the top
> > of the keep, alongside dozens of fellow archers.
>
> I like this idea. Are there any existing RLs in which you can hire/acquire
> computer controlled NPCs?

Try Guild (see guildgame.com)

A.
Anonymous
June 28, 2005 4:12:27 AM

Archived from groups: rec.games.roguelike.development (More info?)

> 3) I want monumental battles. I want to play my small part in some
> epic battle at the gates of Mordor, or defending the people of Rohan in
> Helm's Deep. I want to fight alonside dozens of fellow soldiers, against
> a sea of enemies. I want seige warfare, catapults flinging flaming debris
> and giant boulders, battering rams pounding on the gate, soldiers digging
> a tunnel beneath a section of castle wall in order to collapse it, and
> mighty heroes and villains clashing in the midst of a great battle. I
> want all of it. But, it probably won't happen. Such a game would need to
> make large-scale combat as easy to manage as small-scale one-on-one
> fighting in the dungeons. It's doable, but quite the challenge. I'd love
> to see a game do this though. Nothing would be better than being able to
> take my level 43 Ranger and have him shoot down orc after orc from the top
> of the keep, alongside dozens of fellow archers.

I like this idea. Are there any existing RLs in which you can hire/acquire
computer controlled NPCs? I guess Diablo II does this a little in act 5, but
it's not all that convincing. Any RLs with different factions that fight
each other and that you can join? In any case, an epic battle scene might be
a good demo level for testing some of the stuff I've been putting in my RL.
Related resources
Anonymous
June 28, 2005 9:33:37 AM

Archived from groups: rec.games.roguelike.development (More info?)

Timothy Pruett wrote:
>
> 1) I don't know how many times I've found myself playing a warrior
> class character, and said to myself, "Man, I wish I could get some
> great ammo made, but I can't since I'm not an archer. And I really
> wish I could reforge and beef up my weapons and armor, but I'm not a
> weaponsmith, so I can't do that either." Various classes in games
> like ADoM and ToME have great item creation/modification abilities,
> but those classes are rarely the best-suited to exploit those
> abilities. I always end up wishing I could give my new barbarian some
> of the Arrows of Demon Slaying that my archer had created. But, alas,
> I can't. Well, why not? This is the thought that formed the basis
> for my idea.
>
> Picture this: You start a new game, and get prompted to choose four
> race/class combos. Next, the game world gets generated. Then, you
> choose which of the four characters you want to play as right now.
> All the characters share the same game world, and are affected by the
> same world events. If the mighty Demon of Foo gets killed by one
> character, the Demon of Foo will be dead to all characters in that
> game. At any point in time, the player can save and quit, and play as
> a different character in that game. The player can also exchange
> items between characters in that particular game. For example, the
> player could choose an Alchemist, a Blacksmith, an Archer, and a
> Warrior for his four characters. He can play for a while as all of
> these characters, and, when he feels ready, can prepare his Warrior to
> go for the win, by equipping him with a powerful sword and armor
> crafted by his Blacksmith, enchanted arrows fletched by his Archer,
> and powerful potions brewed by the Alchemist. How cool would that be?
> Basically, it'd be up to the player to choose which four specialized
> classes he wants to use, and then use those specialized abilities to
> prepare one character to go for the win.

So you waant to add Mules to Roguelikes?

Wouldn't it be better just to give the weaponcrafting ability to
barbarians and be done with it?

I always have hated the idea of support characters that are only logged
in to buff up the main character.

What you *want* is to have the best weapons, best ammo, and best
potions, on the best fighter. So why not just allow the "Fighter"
class to get the best weapons, best ammo, and best potions? Saves all
the tedium of logging in and out as different charcters to micromanage
yet another task.

If this were a game like GUILD, I'd have a different comment. However,
you seem to want to use the Weaponsmith, Alchemist, and Archer as
non-adventuring NPCs.

> 2) In every RL out there, the player can count on seeing countless
> items strewn across the various dungeons. Just lying there. In
> populated dungeons. So, my question is, why haven't any of those
> items been picked up by the various dungeon denizens? If I'm walking
> around my apartment complex, I can count on the fact that no item
> found lying in the hallways will stay there for long, because someone
> will inevitably throw it away, or keep it for themselves. Now, random
> items lying about in deserted dungeons makes sense. But these
> dungeons are anything but deserted, and are regularly populated by
> (semi-) intelligent creatures, who would likely find a pile of gold,
> or a glowing dagger, or a magical potion to be definitely worth
> picking up and keeping.

I did this with POWDER in the early days. Items were only generated on
creatures and nothing was left lying around on the floor. My logic was
similar. Surely creatures would pick anything up.

This played *very* poorly. Items lying on the ground serve a very
important role in roguelikes. They provide rewards for exploration
rather than killing. If items are only on creatures, the only way you
gain new items is to kill creatures. There is no longer any strong
reward for avoiding battles. A starting character *has* to fight
everything they see as there is no other way to advance either their
Item set or their XP bar.

If there are items scattered throughout the dungeon, things get much
more interesting. Maybe I should try and avoid that kobold and grab
that wand over there first? Maybe rather than attacking that peaceful
critter I should poke around and see if I can find some better armour
first?

> So, logically, any dungeon that is populated by intelligent creatures
> should have pretty empty hallways. In an enemy fortress, the player
> could probably count on finding weapons and armor all piled in the
> armory, with some extra weapons and such lying about the barracks.
> Food would probably be in a pantry, gold in a treasury, potions and
> herbs in an apothecary or alchemy lab, books and scrolls in a library,
> and so on. This kind of system could lead to some interesting
> gameplay, I think.

"interesting gameplay" has nothing to do with "logical". Entering a
room with a red potion and a scorpion, and deciding whether you should
go for the red potin first as it might be a cure potion which you'll
need when the scorpion poisons you, adds a lot more to the tactics of
the game.

At least, that was my experience with POWDER. Your experience may
differ.

> 3) I want monumental battles. I want to play my small part in some
> epic battle at the gates of Mordor, or defending the people of Rohan
> in Helm's Deep.

I've always liked the idea of this. I've never had a good idea how to
actually implement it. If you make the PC too powerful, he just slays
most of the enemy single handedly and "your" side is window dressing.
If you make the PC the same strength as the rest of the troops, there
will be a huge random element if the PC gets picked on by the enemy AI
and knocked out by a stray catapult.

The battles of Helm's Deep could rely on narrative control to ensure
that one of the giant stone blocks didn't happen to land on a hero
character. If you invoked similar narrative control in the game, the
player will swiftly realize that the giant stone blocks aren't a threat
and ignore them. If you don't, you've got InstaKill attacks, which
most agree aren't fun.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 28, 2005 12:05:06 PM

Archived from groups: rec.games.roguelike.development (More info?)

Chris Morris wrote:
> "Jeff Lait" <torespondisfutile@hotmail.com> writes:
> > Timothy Pruett wrote:
> > > 2) In every RL out there, the player can count on seeing countless
> > > items strewn across the various dungeons. Just lying there. In
> > > populated dungeons. So, my question is, why haven't any of those
> > > items been picked up by the various dungeon denizens? If I'm walking
> > > around my apartment complex, I can count on the fact that no item
> > > found lying in the hallways will stay there for long, because someone
> > > will inevitably throw it away, or keep it for themselves. Now, random
> > > items lying about in deserted dungeons makes sense. But these
> > > dungeons are anything but deserted, and are regularly populated by
> > > (semi-) intelligent creatures, who would likely find a pile of gold,
> > > or a glowing dagger, or a magical potion to be definitely worth
> > > picking up and keeping.
> >
> > This played *very* poorly. Items lying on the ground serve a very
> > important role in roguelikes. They provide rewards for exploration
> > rather than killing. If items are only on creatures, the only way you
> > gain new items is to kill creatures. There is no longer any strong
> > reward for avoiding battles. A starting character *has* to fight
> > everything they see as there is no other way to advance either their
> > Item set or their XP bar.
>
> Well, if the items aren't on the ground because they're on the shelf,
> or the table, or whatever, it's still 'lying around' but not where
> monsters can trip over it. Similarly the healing potions might be in
> the locked medicine cabinet, which you could smash open, lockpick, or
> steal the key from the guard.

Window dressing. What is more, annoying window dressing. Do we get to
play the game of smash every crate? Hoping with each smashed, empty,
crate, for the sword of coolness? Am I the only one with an intense
hatred for the Grass Cutting game mechanic in Zelda?

I'd rather abstract away the crate/medicine chest and let the user see
& react to the loot right away. You can pretend the potions are inside
medicine cabinets and the armour is on armour stands if you wish. I'd
rather save the extra keypresses to do #untrap #unlock #loot on every
item that is spawned in the dungeon.

> > If there are items scattered throughout the dungeon, things get much
> > more interesting. Maybe I should try and avoid that kobold and grab
> > that wand over there first? Maybe rather than attacking that peaceful
> > critter I should poke around and see if I can find some better armour
> > first?
>
> Yes. For deserted dungeons, or only animal dungeons, or even dungeons
> inhabited by intelligent but not especially tidy creatures, items
> scattered around on the floor make sense.

Again, you use "make sense" as if that had any meaning at all. It
doesn't. Dungeon's don't make sense. You can spend hundreds of pages
devising elaborate fanfic to justify why the dungeon is there (which no
one will read). Or you can work on making a fun game.

Tetris is a fun game even though it doesn't "make sense". I don't
doubt one could fashion an elaborate reason why tetris "makes sense".
But one could just as well justify *any* collection of design decisions
in a Roguelike.

> #####
> #?k.# The kobold looks for its wand of disintegration.
> #[%!# The kobold picks up a pair of socks.
> #%/)# The kobold panics!
> #.@!#
>
> #####
> #*k*# The tidy kobold looks on the shelf.
> #.|.# The tidy kobold picks up the wand of disintegration.
> #.|.# The tidy kobold uses the wand of disintegration.
> #.@.# Do you want your possessions identified? [Y/n]

That would be an impressive amount of work on the AI which would be
almost entirely lost on the player. It is almost equivalent to have
the kobold carry said wand with him.

In POWDER, if a kobold stumbles across a wand of fire, he'll pick it
up. He'll then use it on any unsuspecting PC. Of course, there is the
unrealistic generation of all the loot on the ground, allowing the PC
who just entered the dungeon to beat the kobold to the wand, despite
the kobold presumeably living there for a long time. As I mentioned
earlier, this has some very beneficial effects to gameplay.

> > > 3) I want monumental battles. I want to play my small part in some
> > > epic battle at the gates of Mordor, or defending the people of Rohan
> > > in Helm's Deep.
> >
> > I've always liked the idea of this. I've never had a good idea how to
> > actually implement it. If you make the PC too powerful, he just slays
> > most of the enemy single handedly and "your" side is window dressing.
>
> It depends. If there's a wall of troops coming towards where your
> wizard is performing the ritual of Really Big Fireballs, you need to
> hold them off. In most roguelikes 50 grunts are better at this than 1
> strong person. So your side can hold them off while the PC actually
> reduces their numbers and tries to break their ranks.

This has the PC-as-commander. That is very different than playing your
small part in an epic battle. You are playing the decisive part in the
epic battle.

I waas thinking more of the PC being one of the 50 grunts trying to
hold the line against the charge.

> > If you make the PC the same strength as the rest of the troops, there
> > will be a huge random element if the PC gets picked on by the enemy AI
> > and knocked out by a stray catapult.
>
> Once they've got to the hand-to-hand combat they're unlikely to be hit
> by a catapult stone unless the enemy general really doesn't care about
> their forces. While getting there, they need to watch for them coming
> and avoid them, but have a bit of freedom of movement to do so.

If it is so easy to avoid them, why does anyone get crushed?

> > The battles of Helm's Deep could rely on narrative control to ensure
> > that one of the giant stone blocks didn't happen to land on a hero
> > character. If you invoked similar narrative control in the game, the
> > player will swiftly realize that the giant stone blocks aren't a threat
> > and ignore them. If you don't, you've got InstaKill attacks, which
> > most agree aren't fun.
>
> ##### A giant stone block falls from the ceiling!
> .**.F You attempt to dive clear!
> .**.. The fighter is crushed by the stone block.
> A.F@. The archer is crushed by the stone block.
> ..A.. You don't quite make it. [-35 HP] Your armour is badly damaged.
> .A.F. Your potion of healing is shattered. You are stunned.
> #####
>
> It's not (necessarily) an insta-kill, depending on whether you can
> dive out of the way in time, but it's nasty enough that you'd want to
> stay out of the way of them.

I'd have to see the mechanic in action to comment. I would, as a
player, find anything that is an insta-kill (And if "depending on
whether you can dive out of the way" depends on some RNG die roll, it
is still an insta-kill!) very frustrating.

My guess is that you'd have to abandon a lot of roguelike tropes to
pull this off well. I would like to see such a thing done right,
however.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 28, 2005 12:27:46 PM

Archived from groups: rec.games.roguelike.development (More info?)

Jeff Lait wrote:

>> 3) I want monumental battles. I want to play my small part in some
>> epic battle at the gates of Mordor, or defending the people of Rohan
>> in Helm's Deep.

>I've always liked the idea of this. I've never had a good idea how to
>actually implement it.

I think the key to making this fun is in managing the scale. If the
character is a footsoldier in a truly *mass* battle, however powerful
they are, their contribution will always feel either insignificant or
unbalancing. So use text descriptions for the setting, let them know
that the great Siege of Helm's Deep is taking place; then have them
play a key role in a largeish tactical battle, but where the scale is
small enough to be fully understood, such as the retreat into the
tunnels.

--jude hungerford.
Anonymous
June 28, 2005 1:30:54 PM

Archived from groups: rec.games.roguelike.development (More info?)

Chris Morris wrote:
> Of course. Ultimately it has to be fun, but I'd like it to be in some
> way a consistent theme as well, because I think that could make it
> more fun

Yes. The dungeon can be consistent and even realistic, in its own way.
It's surely better option than generic levels without any specific
theme. Themes can be much more than just different looking levels
and with level themes it's possible to build a convincing game world
in the first place.
I think there is no return to the old style with 20+ levels looking
just like the another one.
Anonymous
June 28, 2005 2:34:47 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Shedletsky" <mylastname@stanford.edu> writes:
> > 3) I want monumental battles. I want to play my small part in some
....
> > to see a game do this though. Nothing would be better than being able to
> > take my level 43 Ranger and have him shoot down orc after orc from the top
> > of the keep, alongside dozens of fellow archers.
>
> I like this idea. Are there any existing RLs in which you can hire/acquire
> computer controlled NPCs?

ADOM, Nethack, Crawl (all definitely acquire rather than hire, with
one partial exception from ADOM that I know of)

Oh, and my roguelike had bits of this, and was going to have more (but
that's on hold while I rewrite the engine to be fast enough to cope) -
you could rescue a merchant and his guards from a horde of rats, for example
(though I never quite got that encounter balanced properly, since the
guards were perfectly capable of taking on the rats themselves...)
Not quite an epic battle, but as much of one as would fit on the maps
I was using.

> I guess Diablo II does this a little in act 5, but it's not all that
> convincing. Any RLs with different factions that fight each other
> and that you can join? In any case, an epic battle scene might be a
> good demo level for testing some of the stuff I've been putting in
> my RL.

The faction bits are fairly straightforward to implement - provided a
monster can attack another monster, the rest is easy.

--
Chris
Anonymous
June 28, 2005 2:51:04 PM

Archived from groups: rec.games.roguelike.development (More info?)

Chris Morris wrote:
> "Jeff Lait" <torespondisfutile@hotmail.com> writes:
> > > Well, if the items aren't on the ground because they're on the shelf,
> > > or the table, or whatever, it's still 'lying around' but not where
> > > monsters can trip over it. Similarly the healing potions might be in
> > > the locked medicine cabinet, which you could smash open, lockpick, or
> > > steal the key from the guard.
> >
> > Window dressing. What is more, annoying window dressing. Do we get to
> > play the game of smash every crate? Hoping with each smashed, empty,
> > crate, for the sword of coolness? Am I the only one with an intense
> > hatred for the Grass Cutting game mechanic in Zelda?
>
> If it's on the table or the shelf it can be lying around without
> looking odd - and in plain sight. Depending on the movement nature of
> tables, shelves, etc. that could be made interesting.

Are we still talking about ASCII roguelikes here? Last I checked, you
can only show one glyph per tile. This means you either show a glyph
showing that here is a table there, or you show the glyph that there is
an item there. If you show the item, there is no visual difference
between an item lying on the floor or one on a shelf. The only
difference is the flavour text you use when you go to pick up the item.

Oh, and maybe when you try to bump into the item you knock over the
table you didn't know was there.

> So, you could overturn this table onto the orc and pin them, but that
> would smash the potions on the table, and it might be better to grab
> one and drink it before the orc rounds the table. Or possibly the orc
> doesn't care about the potions and is just going to tip the table onto
> you anyway.

And how, exactly, is this modelled in the ASCII display? Do I see the
table, so know that I will knock it over, but not know that the potion
will be destroyed in the process? Or do I see the item, so
accidentally knock over the table if I don't carefully 'l'ook
everywhere?

> As regards big piles of crates, I'd hope that the monsters labelled
> them (since they don't want to have to open every one when looking for
> the sword either). It might not help if the PC can't read orcish
> hieroglyphs, but I see it as similar to the "a scroll labelled XUG
> IRRDOK" mechanic.
>
> > I'd rather abstract away the crate/medicine chest and let the user see
> > & react to the loot right away. You can pretend the potions are inside
> > medicine cabinets and the armour is on armour stands if you wish. I'd
> > rather save the extra keypresses to do #untrap #unlock #loot on every
> > item that is spawned in the dungeon.
>
> I wouldn't recommend every item be stuck in a chest, but putting the

What were you recommending then? You were decrying having items lying
about on the floor, and said that items should be all found carefully
stowed away in crates and medicine chests.

> occasional item cache in there might be good.

I have no objections to occasional chests with stuff in them. They
need to be rare, however, corresponding with the increased time to deal
with them.

> Perhaps opening the
> crates and disposing of the packing material takes time, so you can
> open the crate of swords and risk the monsters getting through the
> other piles of crates first, or you can just use your makeshift club
> and hope that it's good enough for the fight.

Makeshift club...

In the old Robotech board game there was this cool rule that if you
shot off an opponents arm, one could pick it up and use it as a
makeshift club. Reading it, everyone would go: "Wow, cool..."

It sounds cool to say: "In my roguelike, you can rip the leg off the
table and use it as a makeshift club!!!" But really... What are you
doing in the dungeon without your sword? Oh! I can answer that.
Let's insert another paragraph of text about how this is a baroom
brawl! Or about how the PC was caught and thrown in prison! Or had to
be disarmed to meet an important, but treacherous, merchant!

This all has one common feature. We are taking a hard problem and
replacing it with a harder one. I'm not objecting because I'm
unimaginative. I'm objecting because I'm pragmatic.

> > > Yes. For deserted dungeons, or only animal dungeons, or even dungeons
> > > inhabited by intelligent but not especially tidy creatures, items
> > > scattered around on the floor make sense.
> >
> > Again, you use "make sense" as if that had any meaning at all. It
> > doesn't. Dungeon's don't make sense. You can spend hundreds of pages
> > devising elaborate fanfic to justify why the dungeon is there (which no
> > one will read). Or you can work on making a fun game.
>
> I find it more fun if it looks consistent. The typical roguelike has
> harpies fighting alongside giant lizards fighting alongside kobolds,
> which is fine for a general hack-and-slash game but looks silly.

Well, why don't you go in with an editor and rename the monsters to be
a consistent theme? Just rename them to all different types of lizards
and you have your perfectly consistent Dungeon Of Lizards!

The typical roguelike has the harpies fighting alongside kobolds
because the authors realize that window dressing is less important than
gameplay.

It's easy to say: "Dungeons should be themed!!!" People say it *all*
the time. It seems like half the Angband variants start with efforts
to create one theme or another. And yet themed dungeons are the
exception rather than the norm.

The reason is that it is hard enough to make an unthemed dungeon. It
is hard enough making a roguelike with items scattered on the floor
without having to worry about ensuring each item has a *reason* to be
there.

I don't like these arguments-from-realism because they lead people
astray. People think they need to start off with everything put in the
proper containers. They have to start off with properly meaningful
dungeons. Roguelikes are an abstraction. ASCII based roguelikes are
always going to be a very strong abstraction. Space and time are both
highly discretized. You ignore this at your peril.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 28, 2005 4:00:04 PM

Archived from groups: rec.games.roguelike.development (More info?)

Quoting Timothy Pruett <drakalor.tourist@gmail.com>:
>So, logically, any dungeon that is populated by intelligent creatures
>should have pretty empty hallways.

Logically a dungeon populated by intelligent creatures doesn't make a lot
of sense at all, and specifically would have a rolling shift of four
minotaurs or whatever stationed around the entranceway to beat each novice
adventurer into snot.
--
David Damerell <damerell@chiark.greenend.org.uk> Distortion Field!
Today is First Wednesday, Presuary.
Anonymous
June 28, 2005 4:22:24 PM

Archived from groups: rec.games.roguelike.development (More info?)

David Damerell <damerell@chiark.greenend.org.uk> writes:
> Quoting Timothy Pruett <drakalor.tourist@gmail.com>:
> >So, logically, any dungeon that is populated by intelligent creatures
> >should have pretty empty hallways.
>
> Logically a dungeon populated by intelligent creatures doesn't make a lot
> of sense at all, and specifically would have a rolling shift of four
> minotaurs or whatever stationed around the entranceway to beat each novice
> adventurer into snot.

Sure, for those intelligent creatures able to *afford* however many
minotaurs that costs, and who aren't so weak that employing minotaurs
would be very silly indeed. Where the dungeon as a whole isn't planned
and/or controlled by some superior force, this sort of thing is
unlikely to happen. Sure, the ogres could station a guard on the
entrance, but then the goblins will fill them with arrows the next
time they want to raid the nearby village. So the ogres go off to beat
up the goblins and then no-one is guarding the front door.

Now, a well-guarded entryway of some sort, that's something most
should have, but depending on the dungeon it doesn't have to be at the
start of depth 1.

If it's a goblin lair not too far from civilisation, then they may
well make the first couple of areas empty of goblins and clean of
goblin junk, so that anyone walking in thinks it's just a typical
bat-infested cave system, and then if they do find the concealed
entrance to the actual goblin lair a few 'floors' down, then they get
ambushed and killed. Or not, if the goblins were asleep on their watch
because no-one ever bothers coming this far down.

There's lots of reasons why many dungeons wouldn't have masses of
minotaurs on the front door. Of course, many dungeons *would* and a
wise adventurer would leave those well alone until they were strong
enough.

--
Chris
Anonymous
June 28, 2005 6:28:14 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Jeff Lait" <torespondisfutile@hotmail.com> writes:
> Timothy Pruett wrote:
> > 2) In every RL out there, the player can count on seeing countless
> > items strewn across the various dungeons. Just lying there. In
> > populated dungeons. So, my question is, why haven't any of those
> > items been picked up by the various dungeon denizens? If I'm walking
> > around my apartment complex, I can count on the fact that no item
> > found lying in the hallways will stay there for long, because someone
> > will inevitably throw it away, or keep it for themselves. Now, random
> > items lying about in deserted dungeons makes sense. But these
> > dungeons are anything but deserted, and are regularly populated by
> > (semi-) intelligent creatures, who would likely find a pile of gold,
> > or a glowing dagger, or a magical potion to be definitely worth
> > picking up and keeping.
>
> This played *very* poorly. Items lying on the ground serve a very
> important role in roguelikes. They provide rewards for exploration
> rather than killing. If items are only on creatures, the only way you
> gain new items is to kill creatures. There is no longer any strong
> reward for avoiding battles. A starting character *has* to fight
> everything they see as there is no other way to advance either their
> Item set or their XP bar.

Well, if the items aren't on the ground because they're on the shelf,
or the table, or whatever, it's still 'lying around' but not where
monsters can trip over it. Similarly the healing potions might be in
the locked medicine cabinet, which you could smash open, lockpick, or
steal the key from the guard.

> If there are items scattered throughout the dungeon, things get much
> more interesting. Maybe I should try and avoid that kobold and grab
> that wand over there first? Maybe rather than attacking that peaceful
> critter I should poke around and see if I can find some better armour
> first?

Yes. For deserted dungeons, or only animal dungeons, or even dungeons
inhabited by intelligent but not especially tidy creatures, items
scattered around on the floor make sense.

#####
#?k.# The kobold looks for its wand of disintegration.
#[%!# The kobold picks up a pair of socks.
#%/)# The kobold panics!
#.@!#

#####
#*k*# The tidy kobold looks on the shelf.
#.|.# The tidy kobold picks up the wand of disintegration.
#.|.# The tidy kobold uses the wand of disintegration.
#.@.# Do you want your possessions identified? [Y/n]

> > 3) I want monumental battles. I want to play my small part in some
> > epic battle at the gates of Mordor, or defending the people of Rohan
> > in Helm's Deep.
>
> I've always liked the idea of this. I've never had a good idea how to
> actually implement it. If you make the PC too powerful, he just slays
> most of the enemy single handedly and "your" side is window dressing.

It depends. If there's a wall of troops coming towards where your
wizard is performing the ritual of Really Big Fireballs, you need to
hold them off. In most roguelikes 50 grunts are better at this than 1
strong person. So your side can hold them off while the PC actually
reduces their numbers and tries to break their ranks.

> If you make the PC the same strength as the rest of the troops, there
> will be a huge random element if the PC gets picked on by the enemy AI
> and knocked out by a stray catapult.

Once they've got to the hand-to-hand combat they're unlikely to be hit
by a catapult stone unless the enemy general really doesn't care about
their forces. While getting there, they need to watch for them coming
and avoid them, but have a bit of freedom of movement to do so.

> The battles of Helm's Deep could rely on narrative control to ensure
> that one of the giant stone blocks didn't happen to land on a hero
> character. If you invoked similar narrative control in the game, the
> player will swiftly realize that the giant stone blocks aren't a threat
> and ignore them. If you don't, you've got InstaKill attacks, which
> most agree aren't fun.

##### A giant stone block falls from the ceiling!
..**.F You attempt to dive clear!
..**.. The fighter is crushed by the stone block.
A.F@. The archer is crushed by the stone block.
...A.. You don't quite make it. [-35 HP] Your armour is badly damaged.
..A.F. Your potion of healing is shattered. You are stunned.
#####

It's not (necessarily) an insta-kill, depending on whether you can
dive out of the way in time, but it's nasty enough that you'd want to
stay out of the way of them.

--
Chris
Anonymous
June 28, 2005 6:46:45 PM

Archived from groups: rec.games.roguelike.development (More info?)

Quoting Chris Morris <c.i.morris@durham.ac.uk>:
>David Damerell <damerell@chiark.greenend.org.uk> writes:
>>Quoting Timothy Pruett <drakalor.tourist@gmail.com>:
>>>So, logically, any dungeon that is populated by intelligent creatures
>>>should have pretty empty hallways.
>>Logically a dungeon populated by intelligent creatures doesn't make a lot
>>of sense at all, and specifically would have a rolling shift of four
>>minotaurs or whatever stationed around the entranceway to beat each novice
>>adventurer into snot.
>Sure, for those intelligent creatures able to *afford* however many
>minotaurs that costs, and who aren't so weak that employing minotaurs
>would be very silly indeed.

It doesn't matter what the creatures by the surface think. It's a damn
sight easier for the minotaurs to give up some of their time than to risk
someone getting in, levelling a lot, and killing them all.

Have the archliches station them there if you want some very intelligent
and calculating beastie behind it all.
--
David Damerell <damerell@chiark.greenend.org.uk> Distortion Field!
Today is First Wednesday, Presuary.
Anonymous
June 28, 2005 7:05:00 PM

Archived from groups: rec.games.roguelike.development (More info?)

Jeff Lait wrote:
> I have no objections to occasional chests with stuff in them. They
> need to be rare, however, corresponding with the increased time to deal
> with them.

There doesn't need to be an increased time to deal with items in
unlocked containers (like medicine cabinets).

Suppose a healing potion is in a medicine cabinet. The square's icon is
a medicine cabinet's icon. Looking at it from far away, you see 'A
medicine cabinet'. When you move onto the square, the look text reads:

a) A medicine cabinet
b) A healing potion (in the medicine cabinet)

If you want to pick up the healing potion, just pickup "b".

> Well, why don't you go in with an editor and rename the monsters to be
> a consistent theme? Just rename them to all different types of lizards
> and you have your perfectly consistent Dungeon Of Lizards!
>
> The typical roguelike has the harpies fighting alongside kobolds
> because the authors realize that window dressing is less important than
> gameplay.

Then why do roguelikes bother to have flavour at all? Just to make it
easier to remember which monsters do what? Presumably it would be just
as fun adventuring down to the DIY store to buy wallpaper samples, as
long as the gameplay was identical.

"Window dressing" is a deprecating term for "flavour" or "atmosphere".
And I don't see that atmosphere is less important than gameplay, even
in a very abstracted ASCII-based game.

Roguelikes have the power of the written word at their disposal. Many
of them throw it away in identikit settings and uninteresting messages.
"You attack with your +0 long sword. You hit the giant bat. The giant
bat dies." Yack.

Roguelikes aren't Tetris. They're also not text adventure games. But
they are somewhere in the middle.

Matthew
Anonymous
June 28, 2005 7:39:35 PM

Archived from groups: rec.games.roguelike.development (More info?)

David Damerell <damerell@chiark.greenend.org.uk> writes:
> Quoting Chris Morris <c.i.morris@durham.ac.uk>:
> >David Damerell <damerell@chiark.greenend.org.uk> writes:
> >>Quoting Timothy Pruett <drakalor.tourist@gmail.com>:
> >>>So, logically, any dungeon that is populated by intelligent creatures
> >>>should have pretty empty hallways.
> >>Logically a dungeon populated by intelligent creatures doesn't make a lot
> >>of sense at all, and specifically would have a rolling shift of four
> >>minotaurs or whatever stationed around the entranceway to beat each novice
> >>adventurer into snot.
> >Sure, for those intelligent creatures able to *afford* however many
> >minotaurs that costs, and who aren't so weak that employing minotaurs
> >would be very silly indeed.
>
> It doesn't matter what the creatures by the surface think. It's a damn
> sight easier for the minotaurs to give up some of their time than to risk
> someone getting in, levelling a lot, and killing them all.

In fact, it's probably easier for them to kill or drive off all the
things on the levels above their 'natural' depth and make it a
minotaur dungeon. At that point having high-level (from all the goblin
killing) minotaurs on the entrance makes sense.

> Have the archliches station them there if you want some very intelligent
> and calculating beastie behind it all.

Again a good reason for them to be there.

I wasn't saying that a dungeon shouldn't have tough guards on the
front door, simply that not *every* dungeon inhabited by intelligent
creatures would do, for various reasons.

In a game with multiple dungeons this would make more sense than in a
game with just the one dungeon, because you could have some dungeons
with a heavily guarded front door, and some where the inhabitants just
didn't have the resources to heavily guard it (though they could of
course conceal it, trap it, or do other things).

--
Chris
Anonymous
June 28, 2005 8:47:16 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Jeff Lait" <torespondisfutile@hotmail.com> writes:
> > Well, if the items aren't on the ground because they're on the shelf,
> > or the table, or whatever, it's still 'lying around' but not where
> > monsters can trip over it. Similarly the healing potions might be in
> > the locked medicine cabinet, which you could smash open, lockpick, or
> > steal the key from the guard.
>
> Window dressing. What is more, annoying window dressing. Do we get to
> play the game of smash every crate? Hoping with each smashed, empty,
> crate, for the sword of coolness? Am I the only one with an intense
> hatred for the Grass Cutting game mechanic in Zelda?

If it's on the table or the shelf it can be lying around without
looking odd - and in plain sight. Depending on the movement nature of
tables, shelves, etc. that could be made interesting.

So, you could overturn this table onto the orc and pin them, but that
would smash the potions on the table, and it might be better to grab
one and drink it before the orc rounds the table. Or possibly the orc
doesn't care about the potions and is just going to tip the table onto
you anyway.

As regards big piles of crates, I'd hope that the monsters labelled
them (since they don't want to have to open every one when looking for
the sword either). It might not help if the PC can't read orcish
hieroglyphs, but I see it as similar to the "a scroll labelled XUG
IRRDOK" mechanic.

> I'd rather abstract away the crate/medicine chest and let the user see
> & react to the loot right away. You can pretend the potions are inside
> medicine cabinets and the armour is on armour stands if you wish. I'd
> rather save the extra keypresses to do #untrap #unlock #loot on every
> item that is spawned in the dungeon.

I wouldn't recommend every item be stuck in a chest, but putting the
occasional item cache in there might be good. Perhaps opening the
crates and disposing of the packing material takes time, so you can
open the crate of swords and risk the monsters getting through the
other piles of crates first, or you can just use your makeshift club
and hope that it's good enough for the fight.

> > Yes. For deserted dungeons, or only animal dungeons, or even dungeons
> > inhabited by intelligent but not especially tidy creatures, items
> > scattered around on the floor make sense.
>
> Again, you use "make sense" as if that had any meaning at all. It
> doesn't. Dungeon's don't make sense. You can spend hundreds of pages
> devising elaborate fanfic to justify why the dungeon is there (which no
> one will read). Or you can work on making a fun game.

I find it more fun if it looks consistent. The typical roguelike has
harpies fighting alongside giant lizards fighting alongside kobolds,
which is fine for a general hack-and-slash game but looks silly.

I'd rather the giant lizards were in one dungeon and the kobolds
(except as stripped-clean skeletons) were in another. That way the
level generation, item generation, and creature strategy can fit with
the type of creatures.

So in the hurriedly abandoned mines, there's mining gear just
scattered where people dropped it as they ran, whereas the working
mines have it either carried by the miners or in neat piles in the
quartermaster's office.

If it takes a hundred pages to explain the logic behind the dungeon
it's not worth it. If it takes a page of notes on the theme then it
is, IMO.

> Tetris is a fun game even though it doesn't "make sense". I don't
> doubt one could fashion an elaborate reason why tetris "makes sense".
> But one could just as well justify *any* collection of design decisions
> in a Roguelike.

Of course. Ultimately it has to be fun, but I'd like it to be in some
way a consistent theme as well, because I think that could make it
more fun (compare places like the Tower of Eternal Flames in ADoM, or
the dungeon branches in Crawl, to the main generic dungeons in those
games).

> > It depends. If there's a wall of troops coming towards where your
> > wizard is performing the ritual of Really Big Fireballs, you need to
> > hold them off. In most roguelikes 50 grunts are better at this than 1
> > strong person. So your side can hold them off while the PC actually
> > reduces their numbers and tries to break their ranks.
>
> This has the PC-as-commander. That is very different than playing your
> small part in an epic battle. You are playing the decisive part in the
> epic battle.

True, but if the PC's actions can't be the deciding factor, why can't
the PC just go to the back of the grunts and pick off the occasional
thing that comes through? Or walk off entirely and have a drink? Much
safer and will get the same result overall.

> I waas thinking more of the PC being one of the 50 grunts trying to
> hold the line against the charge.

Balancing this to make it interesting is tricky, though. I suppose
from the PC's point of view it doesn't matter what's going on outside
their FOV, so just abstract all the rest away and the PC's task is to
stay alive until the end of the battle, and then either loot the
bodies or run away in defeat.

> > > If you make the PC the same strength as the rest of the troops, there
> > > will be a huge random element if the PC gets picked on by the enemy AI
> > > and knocked out by a stray catapult.
> >
> > Once they've got to the hand-to-hand combat they're unlikely to be hit
> > by a catapult stone unless the enemy general really doesn't care about
> > their forces. While getting there, they need to watch for them coming
> > and avoid them, but have a bit of freedom of movement to do so.
>
> If it is so easy to avoid them, why does anyone get crushed?

I thought catapults were mainly siege weapons anyway. Difficult to aim
by turning, slow to load, lots of damage against anything too slow to
get out of the way. A loose formation of infantry should be largely okay.

Well, either siege or attacking troops with little freedom of movement.

Plus they don't work underground, or particularly well in forests,
which gives plenty of venues for epic battles where things three
screens away aren't an immediate concern for the PC.

> > ##### A giant stone block falls from the ceiling!
> > .**.F You attempt to dive clear!
> > .**.. The fighter is crushed by the stone block.
> > A.F@. The archer is crushed by the stone block.
> > ..A.. You don't quite make it. [-35 HP] Your armour is badly damaged.
> > .A.F. Your potion of healing is shattered. You are stunned.
> > #####
> >
> > It's not (necessarily) an insta-kill, depending on whether you can
> > dive out of the way in time, but it's nasty enough that you'd want to
> > stay out of the way of them.
>
> I'd have to see the mechanic in action to comment. I would, as a
> player, find anything that is an insta-kill (And if "depending on
> whether you can dive out of the way" depends on some RNG die roll, it
> is still an insta-kill!) very frustrating.

If you know where the stone blocks that _might_ fall are, you can
avoid them if you pick your movement right (though do you back off
under that shaky-looking block or stay where you are and take more
hits from the ogre?) - in which case it's just a calculated risk to
step under it, and so a possible instakill is fine. I'd never make it
an instakill, even if you completely failed the dive test, but it
would probably take more HP than you have to survive it.

If you don't know where the stone blocks that might fall are, then
having one land on you would be very annoying indeed, unless there was
some other precaution you could take to set them off in advance of you
or hold them up until you passed.

--
Chris
Anonymous
June 29, 2005 3:55:05 AM

Archived from groups: rec.games.roguelike.development (More info?)

In article <87hdfiqy9n.fsf@dinopsis.dur.ac.uk>, Chris Morris <c.i.morris@durham.ac.uk> wrote:

>True, but if the PC's actions can't be the deciding factor, why can't
>the PC just go to the back of the grunts and pick off the occasional
>thing that comes through? Or walk off entirely and have a drink? Much
>safer and will get the same result overall.

Have troop morale be proportional to the NPC's distance from the
player.

Alan
Anonymous
June 29, 2005 11:00:01 AM

Archived from groups: rec.games.roguelike.development (More info?)

Timothy Pruett wrote:

> 3) I want monumental battles.

>From a reality POV, it makes sense for most fighters to be or have been
soldiers, and hence, in a role-playing game, to act in concert with an
army.
Anonymous
June 29, 2005 11:31:52 AM

Archived from groups: rec.games.roguelike.development (More info?)

Matthew Bennett wrote:
> Jeff Lait wrote:
> > I have no objections to occasional chests with stuff in them. They
> > need to be rare, however, corresponding with the increased time to deal
> > with them.
>
> There doesn't need to be an increased time to deal with items in
> unlocked containers (like medicine cabinets).
>
> Suppose a healing potion is in a medicine cabinet. The square's icon is
> a medicine cabinet's icon. Looking at it from far away, you see 'A
> medicine cabinet'. When you move onto the square, the look text reads:
>
> a) A medicine cabinet
> b) A healing potion (in the medicine cabinet)
>
> If you want to pick up the healing potion, just pickup "b".

While this addresses the interface problems of looting the cabinet, it
still requires the player to walk over every cabinet to see that it is
empty. When I return to the room later, I get to again walk over them
to double check they are empty. Or maybe use the OCD most roguelike
players develop to methodically smash every cabinet.

Modelling junk is always a concern for me. I'm very much in the
Nethack school of thought: The dungeon *has* bones, medicine cabinets,
whatever, they just aren't shown if they aren't important.

> > Well, why don't you go in with an editor and rename the monsters to be
> > a consistent theme? Just rename them to all different types of lizards
> > and you have your perfectly consistent Dungeon Of Lizards!
> >
> > The typical roguelike has the harpies fighting alongside kobolds
> > because the authors realize that window dressing is less important than
> > gameplay.
>
> Then why do roguelikes bother to have flavour at all?

Because less important doesn't mean unimportant.

> Just to make it
> easier to remember which monsters do what?

Yes and no. To make it easier to visualize the monsters. If you want
20 distinct monsters, you are going to have troubles if everything is a
different shaped kobold. By spreading the theme net wide you can
easily arrive at a few dozen different monsters that people will be
able to distinguish.

I'm also not at all convinced that it is any more unrealistic to have
the dungeon of giant lizards than it is to have the dungeon of mixed
foes. I've engaged my suspension of disbelief circuits right when we
got dumped in a mysterious dungeon of enemies.

To make a really consistent and good dungeon requires care, and, IMHO,
human intervention. This would be an RPG, not a roguelike. I like to
claim there is a difference between roguelikes and RPGs.

> Presumably it would be just
> as fun adventuring down to the DIY store to buy wallpaper samples, as
> long as the gameplay was identical.

Yes! It would! This would be an example of a nicely strongly themed
dungeon, right?

> "Window dressing" is a deprecating term for "flavour" or "atmosphere".
> And I don't see that atmosphere is less important than gameplay, even
> in a very abstracted ASCII-based game.

Flavour and atmosphere becomes repetitive *very* quickly if you don't
generate it well. There's only so many medicine cabinets you need to
smash before they blur into just another loot box. Most roguelikes
have long play cycles, so I have little patience for flavour that
fades.

Good flavour must be varied. Which is hard.

> Roguelikes have the power of the written word at their disposal. Many
> of them throw it away in identikit settings and uninteresting messages.
> "You attack with your +0 long sword. You hit the giant bat. The giant
> bat dies." Yack.

If they correctly conjugate the verb, I consider it a good start.

How do you generate your interesting messages if you cannot yet
generate the boring ones? It isn't easy generating the boring
messages. It isn't any easier generating the interseting ones. The hard
part about roguelikes is data entry. And interesting messages compound
data entry. I didn't want "The giant bat dies!" for all my death
messages. So, I added stuff like: "The giant bat collapses in a pool
of blood!" Of course, then skeletons can't very well collapse in a
pool of blood, so I needed to seven different sets of death messages to
cover all the different classes of critters.

> Roguelikes aren't Tetris. They're also not text adventure games. But
> they are somewhere in the middle.

They are very far removed from text adventure games. Text adventure
games can rely on prebuilt static text.

I tried something close to text adventures with You Only Live Once
(http://www.zincland.com/7drl/liveonce) It is close to a text
adventure in terms of having lots of prebuilt descriptions.
Replayability, however, suffers.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 29, 2005 12:00:44 PM

Archived from groups: rec.games.roguelike.development (More info?)

"jude hungerford" <lichen678@popmail.com> wrote in message
news:1119972466.471244.272970@f14g2000cwb.googlegroups.com...
[...]
> I think the key to making this fun is in managing the scale. If the
> character is a footsoldier in a truly *mass* battle, however powerful
> they are, their contribution will always feel either insignificant or
> unbalancing. So use text descriptions for the setting, let them know
> that the great Siege of Helm's Deep is taking place; then have them
> play a key role in a largeish tactical battle, but where the scale is
> small enough to be fully understood, such as the retreat into the
> tunnels.
[...]

Did you play Neverwinter Nights: Hordes of the Underdark? One of my favorite
scenes in that expansion (which totally redeemed a mediocre series, might I
add) was the "epic battle". Fun stuff.

I really most people in this thread are overly thinking this or overly
engineering a very simple concept and implementation. Catapults giving you
trouble? Take them out for now, they're not *required*.

All it takes to pull off something like this is:
1) Generating extra enemies (more that the player can handle alone)
2) Generating allied NPCs that can fight those enemies.

You do this in a large enough scale, with good enough atmosphere, detail,
and diversity, and you've got yourself a pretty interesting epic battle
scene in an RL.

Diablo II in Act 5 satisfied #2 with the barbarians, but not #1-- these
allied NPCs were just for flavor, they didn't really compare to you in
power, and they really weren't doing much in holding off the enemies, which,
additionally, were easily handled by the player. Maybe if the barbarians
would've been 5 times as strong, 3 times as many, and the monsters 5 times
as many, then we'd be talking. At this point 1) The player can really use
the extra help with the extra enemies, and 2) The NPCs can actually provide
that help. Granted, all this is written from a PC-centric point of view,
because that is the tradition of RL (or any game) design, but in essense the
PC could be just as powerful/significant as the other allied NPCs or even
less.


--Nolithius
Anonymous
June 29, 2005 12:01:28 PM

Archived from groups: rec.games.roguelike.development (More info?)

Chris Morris wrote:
> "Jeff Lait" <torespondisfutile@hotmail.com> writes:
>
> > It's easy to say: "Dungeons should be themed!!!" People say it *all*
> > the time. It seems like half the Angband variants start with efforts
> > to create one theme or another. And yet themed dungeons are the
> > exception rather than the norm.
>
> You mean Angband *isn't* centipede themed?
>
> Crawl has many themed side-branches which work well for giving
> different tactical challenges. ADoM has this to an extent (ToEF,
> Minotaur Maze, various special levels in the CoC, etc). Nethack
> similarly (Gnomish Mines, etc).

I don't know about the development history of Crawl. However, Nethack
and ADOM both are notable in adding the themed dungeons *after*
building the unthemed dungeon.

I think a lot of people here for get ADOM once stood for Ancient
Dungeon of Mystery as there was only one dungeon and no surface world.

> Now, they all also have an unthemed trunk, but I don't think it's
> impossible to reduce the importance of that bit, or remove it
> entirely. I'm not suggesting making a variant of any of the above with
> the trunk removed - but I don't think it'd be particularly difficult
> to make a fun game where most of the levels had some theme.

Well, if you think writing a brand new version of Crawl isn't
"particularly difficult" then you are right, I guess. I believe there
was a group that was looking at a version 5 rewrite of Crawl that
stalled. Maybe you could just quickly finish it up for them?

> > The reason is that it is hard enough to make an unthemed dungeon. It
> > is hard enough making a roguelike with items scattered on the floor
> > without having to worry about ensuring each item has a *reason* to be
> > there.
>
> *shrug* It doesn't have to be that much harder.

I'm just bitter because after a few years I still haven't got to this
"not much harder" part of roguelike development.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 29, 2005 12:06:24 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Jeff Lait" <torespondisfutile@hotmail.com> wrote in message
news:1119981064.427499.223460@z14g2000cwz.googlegroups.com...
[...]
> Are we still talking about ASCII roguelikes here? Last I checked, you
> can only show one glyph per tile. This means you either show a glyph
> showing that here is a table there, or you show the glyph that there is
> an item there. If you show the item, there is no visual difference
> between an item lying on the floor or one on a shelf. The only
> difference is the flavour text you use when you go to pick up the item.
[...]

I really dislike the idea as much as you do, Jeff, and agree with your
points about abstraction-- but uh... a brown background can be a table. Just
wanted to point that out ;) . But yeah, ultimately, in RLs it's pretty
useless to try to beat the granularity to death in this fashion, the game
would ultimately not improve one bit. Now, in graphical (specifically 3D)
games like Morrowind, "containers" such as tables or armoires with items on
top do look very cool ;) .

--Nolithius
Anonymous
June 29, 2005 12:28:23 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Timothy Pruett" <drakalor.tourist@gmail.com> wrote in message
news:E-idnTBTTad-913fRVn-rA@adelphia.com...
[...]
> 1) I don't know how many times I've found myself playing a warrior
> class character, and said to myself, "Man, I wish I could get some
> great ammo made, but I can't since I'm not an archer. And I really
> wish I could reforge and beef up my weapons and armor, but I'm not a
> weaponsmith, so I can't do that either." Various classes in games
> like ADoM and ToME have great item creation/modification abilities,
> but those classes are rarely the best-suited to exploit those
> abilities. I always end up wishing I could give my new barbarian some
> of the Arrows of Demon Slaying that my archer had created. But, alas,
> I can't. Well, why not? This is the thought that formed the basis
> for my idea.
>
> Picture this: You start a new game, and get prompted to choose four
> race/class combos. Next, the game world gets generated. Then, you
> choose which of the four characters you want to play as right now.
> All the characters share the same game world, and are affected by the
> same world events. If the mighty Demon of Foo gets killed by one
> character, the Demon of Foo will be dead to all characters in that
> game. At any point in time, the player can save and quit, and play as
> a different character in that game. The player can also exchange
> items between characters in that particular game. For example, the
> player could choose an Alchemist, a Blacksmith, an Archer, and a
> Warrior for his four characters. He can play for a while as all of
> these characters, and, when he feels ready, can prepare his Warrior to
> go for the win, by equipping him with a powerful sword and armor
> crafted by his Blacksmith, enchanted arrows fletched by his Archer,
> and powerful potions brewed by the Alchemist. How cool would that be?
> Basically, it'd be up to the player to choose which four specialized
> classes he wants to use, and then use those specialized abilities to
> prepare one character to go for the win.
>
> Granted, for this idea to be fun, all the classes would have to be fun
> to play as, and the item creation/modification would have to be fun as
> well, otherwise the game would boil down to a tedious process of scumming.
[...]

This is, at best, poor design, and, at worst, easily exploitable mechanics
that can totally break your game.

Basically you're talking about having mules or alternate support characters
whose sole purpose is to create items for the main character-- you're
talking about actually making a game that encourages this-- this is poor
design.

In the end something like this would be so tedious that it would break the
game. Why not just have a system of skill selection under which you can
chose craft skills? Why not make your game classless and totally
skill-based, so the player can pick up whatever he wants as he needs to?

The reason some games chose to abstract this flexibility away is to
*balance* the classes. Yes, crafting classes aren't the best fighters, but
they can craft :p . Fighting classes can't craft their armor, but can fight
well. Wizards can't fight well, but can cast magic. This is just a classical
balance-driven design, where certain classes (or other choices available to
the player) have pros and cons which ideally fit a certain playing style or
desired effect. This is a conscious abstraction from the flexibility of an
openly skill-based design because 1) it is easy to design and implement, and
2) it allows you to design interesting tactical challenges for the player
based solely on the predetermined strengths and weaknesses of his choices.
For example, you can design an area/monster/item that you'd want Warriors to
have a hard time with, but Wizards an easy time, etc. etc. This can be done
for plot purposes or just for fun. This approach is, on the whole, just
easier to deal with from a design and development point of view, whether it
is "better" or "worse" depends on personal taste.

Although making openly skill-based games can actually give the player a lot
of flexibility and create a lot of fun (such as coming up with effective
combinations of skills, etc), it poses additional challenges in the design
of the game's plot/areas/monster/items/etc.. If these challenges aren't
overcome successfully from a developer's perspective the player will end up
with a game that is very flexible but has very little else to offer, or has
inconsistant difficulty, or waves in and out between fun and tedium.

It's really up to the developer in the end. I myself favor entirely
skill-based mechanics, but I am not naive as to the implications of such a
system. This is why I respect class-based systems that add additional
abstraction for the sake of sanity.

--Nolithius
Anonymous
June 29, 2005 12:49:15 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Timothy Pruett" wrote:

[snip]
>
> 3) I want monumental battles. I want to play my small part in some
> epic battle at the gates of Mordor, or defending the people of Rohan
> in Helm's Deep. I want to fight alonside dozens of fellow soldiers,
> against a sea of enemies. I want seige warfare, catapults flinging
> flaming debris and giant boulders, battering rams pounding on the
> gate, soldiers digging a tunnel beneath a section of castle wall in
> order to collapse it, and mighty heroes and villains clashing in the
> midst of a great battle. I want all of it. But, it probably won't
> happen. Such a game would need to make large-scale combat as easy to
> manage as small-scale one-on-one fighting in the dungeons. It's
> doable, but quite the challenge. I'd love to see a game do this
> though. Nothing would be better than being able to take my level 43
> Ranger and have him shoot down orc after orc from the top of the keep,
> alongside dozens of fellow archers.

As far as epic battles where the pc is involved personally, I don't think
rls are well-suited. But a pc as an overall commander.. that might work.

In my BDR, I have creatures organised into Squads. Each squad has a
Sergeant/Commander who is responsible for ordering the squad members.
Each Commander is then organised into more squads... forming a virtual
army. (The idea behind this was to encourage "team play" by having
parties, rather than just a single pc. I like Antoine's direction in
Guild, but I'm thinking less management required, and having the rest
of the party be more like Nethack "pets", but with more orders)

One idea I have would be to simulate epic battles by having the pc give
orders to individual Commanders, who would then pass the orders/strategy
onto its team members.

To reduce micro-management (so the computer AI, which never gets bored,
wouldn't have an unfair advantage), the player/AI can only give orders to
one squad per 5 or so turns.

The pc is not present on the playmap at all, rather overseeing the battle
as a whole. If you've ever played the PC games "Sacrifice", imagine the
pc to be an incorporeal wizard.

I was thinking perhaps this wouldn't be a bad little minigame as part
of the "levelling-up" process for pcs, say, in order to obtain the
special "leadership" ability. Not an often occurrence, but just every
now and then as the pc's situation requires.

Cheers,
Adam
Anonymous
June 29, 2005 1:06:56 PM

Archived from groups: rec.games.roguelike.development (More info?)

Jeff Lait wrote:
> While this addresses the interface problems of looting the cabinet, it
> still requires the player to walk over every cabinet to see that it is
> empty. When I return to the room later, I get to again walk over them
> to double check they are empty. Or maybe use the OCD most roguelike
> players develop to methodically smash every cabinet.

OK, then let empty, explored medicine cabinets display no icon. If you
want it to be completely transparent and identical in every way to
there being no medicine cabinet, you could display a line of text in
that square ("There is a medicine cabinet here"), but show the potion's
icon even from far away. That would still be an improvement.

> Modelling junk is always a concern for me. I'm very much in the
> Nethack school of thought: The dungeon *has* bones, medicine cabinets,
> whatever, they just aren't shown if they aren't important.

Maybe. But Nethack doesn't do this very consistently. It has grafitti
("Vlad was here") that isn't important. Imps taunt you, which isn't
important (unless you're blind or hallucinating, and then not very
important). Fortune cookies have messages in them, which aren't
important.

If a game models sinks, toilets, chests, thrones and doors, it's
describing things at a certain level of detail. Anything at that level
which it doesn't describe (chairs, tables, medicine cabinets), I tend
not to imagine. In what sense does the dungeon *have* these things if
they aren't visible in any way?

> I'm also not at all convinced that it is any more unrealistic to have
> the dungeon of giant lizards than it is to have the dungeon of mixed
> foes. I've engaged my suspension of disbelief circuits right when we
> got dumped in a mysterious dungeon of enemies.

By that token, since the player's suspension of disbelief circuits are
engaged, you could insert absolutely any kind of monster without
hurting the game. "The ravioli curses you. The Michael Keaton hits you
with its +2 nelson's column."

> To make a really consistent and good dungeon requires care, and, IMHO,
> human intervention. This would be an RPG, not a roguelike. I like to
> claim there is a difference between roguelikes and RPGs.
>
> > Presumably it would be just
> > as fun adventuring down to the DIY store to buy wallpaper samples, as
> > long as the gameplay was identical.
>
> Yes! It would! This would be an example of a nicely strongly themed
> dungeon, right?

Strongly themed, yes. I don't think I would like it though. Killing
harpies/kobolds/whatever is at least a bit more fun than selecting
wallpaper.

I'm not arguing for themed dungeons per se. I'm arguing, in a
roundabout way, that consistency (among other things) tends to increase
immersion, which tends to increase enjoyment.

> Flavour and atmosphere becomes repetitive *very* quickly if you don't
> generate it well. There's only so many medicine cabinets you need to
> smash before they blur into just another loot box. Most roguelikes
> have long play cycles, so I have little patience for flavour that
> fades.

There are only so many goblins you can generate before they blur into
just another goblin (the number is around 3). There are only so many
berserk rages you can fly into and kill a party of orcs in before it's
just another berserk rage and just another party of orcs. There are
only so many times you can rub a lamp and wish for a GDSM, etc.

The novelty of everything wears off eventually. But I would rather play
a Nethack where there are fortune cookies/graffiti/imp taunts. Even
when the novelty is gone, the cumulative effect of small amounts of
flavour is large.

> > Roguelikes have the power of the written word at their disposal. Many
> > of them throw it away in identikit settings and uninteresting messages.
> > "You attack with your +0 long sword. You hit the giant bat. The giant
> > bat dies." Yack.
>
> If they correctly conjugate the verb, I consider it a good start.
> How do you generate your interesting messages if you cannot yet
> generate the boring ones?

Which verb did I conjugate incorrectly?

> I tried something close to text adventures with You Only Live Once
> (http://www.zincland.com/7drl/liveonce) It is close to a text
> adventure in terms of having lots of prebuilt descriptions.
> Replayability, however, suffers.

I quite liked YOLO. Actually, I played it a couple of weeks ago, and I
thought the text made it significantly more immersive and fun.
Replayability suffers, but no 7DRL is going to be amazingly replayable,
is it?

(The other 7DRL I've played, the Sheep's Z-DAY, has virtually no
textual description, and is very light on gameplay mechanics, but still
has a lot of atmosphere, and I got a lot of enjoyment out of it for
exactly that reason. "Zombie apocalypse" is one of my favourite
settings.)

Matthew
Anonymous
June 29, 2005 1:33:50 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Jeff Lait" <torespondisfutile@hotmail.com> writes:
> Chris Morris wrote:
> > If it's on the table or the shelf it can be lying around without
> > looking odd - and in plain sight. Depending on the movement nature of
> > tables, shelves, etc. that could be made interesting.
>
> Are we still talking about ASCII roguelikes here? Last I checked, you

I'm not sure we were ever talking about solely ASCII roguelikes here;
I must have missed that bit. I admit I'd be unlikely to do a non-ASCII
one, but then I'd be unlikely to implement this entire thread in that
case.

> can only show one glyph per tile. This means you either show a glyph
> showing that here is a table there, or you show the glyph that there is
> an item there. If you show the item, there is no visual difference
> between an item lying on the floor or one on a shelf. The only
> difference is the flavour text you use when you go to pick up the item.
>
> Oh, and maybe when you try to bump into the item you knock over the
> table you didn't know was there.

Well, that would just be bad interface design if the automatic action
towards tables was 'tip'.

> > So, you could overturn this table onto the orc and pin them, but that
> > would smash the potions on the table, and it might be better to grab
> > one and drink it before the orc rounds the table. Or possibly the orc
> > doesn't care about the potions and is just going to tip the table onto
> > you anyway.
>
> And how, exactly, is this modelled in the ASCII display? Do I see the
> table, so know that I will knock it over, but not know that the potion
> will be destroyed in the process? Or do I see the item, so
> accidentally knock over the table if I don't carefully 'l'ook
> everywhere?

It wouldn't be suited to traditional ASCII. It'd obviously be okay
with tiles, but it might also work with the zoomed ASCII suggested
previously on the group.
+--+--+--+
|.@|**|.o|
|..|!*|..|
+--+--+--+

Did anyone ever try zoomed ASCII? I've no idea how well it would work
in practice. Possibly it'd be best just to go to tiles at that stage.

> > I wouldn't recommend every item be stuck in a chest, but putting the
>
> What were you recommending then? You were decrying having items lying
> about on the floor, and said that items should be all found carefully
> stowed away in crates and medicine chests.

I didn't say *all* found.

> It sounds cool to say: "In my roguelike, you can rip the leg off the
> table and use it as a makeshift club!!!" But really... What are you
> doing in the dungeon without your sword? Oh! I can answer that.
> Let's insert another paragraph of text about how this is a baroom
> brawl! Or about how the PC was caught and thrown in prison! Or had to
> be disarmed to meet an important, but treacherous, merchant!

Well, a bar-room brawl or prison break might be a quite good subject
for an nDRL.

> > I find it more fun if it looks consistent. The typical roguelike has
> > harpies fighting alongside giant lizards fighting alongside kobolds,
> > which is fine for a general hack-and-slash game but looks silly.
>
> Well, why don't you go in with an editor and rename the monsters to be
> a consistent theme? Just rename them to all different types of lizards
> and you have your perfectly consistent Dungeon Of Lizards!

Well, because it doesn't matter in every game.

> The typical roguelike has the harpies fighting alongside kobolds
> because the authors realize that window dressing is less important than
> gameplay.

I said it was fine for a general hack-and-slash game. Not every
roguelike has to be like that.

> It's easy to say: "Dungeons should be themed!!!" People say it *all*
> the time. It seems like half the Angband variants start with efforts
> to create one theme or another. And yet themed dungeons are the
> exception rather than the norm.

You mean Angband *isn't* centipede themed?

Crawl has many themed side-branches which work well for giving
different tactical challenges. ADoM has this to an extent (ToEF,
Minotaur Maze, various special levels in the CoC, etc). Nethack
similarly (Gnomish Mines, etc).

Now, they all also have an unthemed trunk, but I don't think it's
impossible to reduce the importance of that bit, or remove it
entirely. I'm not suggesting making a variant of any of the above with
the trunk removed - but I don't think it'd be particularly difficult
to make a fun game where most of the levels had some theme.

> The reason is that it is hard enough to make an unthemed dungeon. It
> is hard enough making a roguelike with items scattered on the floor
> without having to worry about ensuring each item has a *reason* to be
> there.

*shrug* It doesn't have to be that much harder.

--
Chris
Anonymous
June 29, 2005 1:38:30 PM

Archived from groups: rec.games.roguelike.development (More info?)

Matthew Bennett wrote:
> Jeff Lait wrote:
> > While this addresses the interface problems of looting the cabinet, it
> > still requires the player to walk over every cabinet to see that it is
> > empty. When I return to the room later, I get to again walk over them
> > to double check they are empty. Or maybe use the OCD most roguelike
> > players develop to methodically smash every cabinet.
>
> OK, then let empty, explored medicine cabinets display no icon. If you
> want it to be completely transparent and identical in every way to
> there being no medicine cabinet, you could display a line of text in
> that square ("There is a medicine cabinet here"), but show the potion's
> icon even from far away. That would still be an improvement.

Agreed, it would be an improvement, provided the medicine cabinet
itself had a reason to be there.

Otherwise, people will be asking: "Why are there these medicine
cabinets in all of these rooms?" rather than "Why are these potions
scattered about?"

> > Modelling junk is always a concern for me. I'm very much in the
> > Nethack school of thought: The dungeon *has* bones, medicine cabinets,
> > whatever, they just aren't shown if they aren't important.
>
> Maybe. But Nethack doesn't do this very consistently. It has grafitti
> ("Vlad was here") that isn't important. Imps taunt you, which isn't
> important (unless you're blind or hallucinating, and then not very
> important). Fortune cookies have messages in them, which aren't
> important.

All of those are rare events. They are not on the scale of every room
or every potion.

Here's a litmus test for you: If you ran into a medicine cabinet in
Nethack, what would you do? You wouldn't think it was just a text
description to justify the existence of a potion.

Conversely, if every potion was properly encapsulated in text-only
medicine cabinets, what would you do when you encounter a sink? Say:
"Ah, more window dressing!" and move on? Instead, the sink stands out
and demands investigation.

The same problem happened to graphical adventure games when their
graphics got too good. Suddenly the active objects blended seemlessly
into the background, turning the game into a "Click everywhere on the
screen" challenge. One particularly painful Kings Quest had a library
where each book provided different cute little fake book titles. I
grew bored after a half dozen, so got stuck because behind one of the
books was a slip of paper you needed to progress.

> If a game models sinks, toilets, chests, thrones and doors, it's
> describing things at a certain level of detail. Anything at that level
> which it doesn't describe (chairs, tables, medicine cabinets), I tend
> not to imagine. In what sense does the dungeon *have* these things if
> they aren't visible in any way?

In the sense that it is not a roleplaying game. It is a tactical
combat game. Roguelikes are not CRPGs.

> > I'm also not at all convinced that it is any more unrealistic to have
> > the dungeon of giant lizards than it is to have the dungeon of mixed
> > foes. I've engaged my suspension of disbelief circuits right when we
> > got dumped in a mysterious dungeon of enemies.
>
> By that token, since the player's suspension of disbelief circuits are
> engaged, you could insert absolutely any kind of monster without
> hurting the game. "The ravioli curses you. The Michael Keaton hits you
> with its +2 nelson's column."

I think the answer to that is "Yes". NetHack seems to be doing quite
well with its silly choices. I see enough references to Zangband's
"Silly" monsters to guess that NetHack isn't alone.

> > To make a really consistent and good dungeon requires care, and, IMHO,
> > human intervention. This would be an RPG, not a roguelike. I like to
> > claim there is a difference between roguelikes and RPGs.
> >
> > > Presumably it would be just
> > > as fun adventuring down to the DIY store to buy wallpaper samples, as
> > > long as the gameplay was identical.
> >
> > Yes! It would! This would be an example of a nicely strongly themed
> > dungeon, right?
>
> Strongly themed, yes. I don't think I would like it though. Killing
> harpies/kobolds/whatever is at least a bit more fun than selecting
> wallpaper.

If it uses the same mechanic, I strongly suspect I'd prefer wallpaper
selection. It would at least seem fresh and exciting until I realized
that the underlying system is the same.

> I'm not arguing for themed dungeons per se. I'm arguing, in a
> roundabout way, that consistency (among other things) tends to increase
> immersion, which tends to increase enjoyment.

I'm all for consistency. I try to work very hard to make the POWDER
world consistent. I've been trying to avoid silly monsters.

> > Flavour and atmosphere becomes repetitive *very* quickly if you don't
> > generate it well. There's only so many medicine cabinets you need to
> > smash before they blur into just another loot box. Most roguelikes
> > have long play cycles, so I have little patience for flavour that
> > fades.
>
> There are only so many goblins you can generate before they blur into
> just another goblin (the number is around 3). There are only so many
> berserk rages you can fly into and kill a party of orcs in before it's
> just another berserk rage and just another party of orcs. There are
> only so many times you can rub a lamp and wish for a GDSM, etc.
>
> The novelty of everything wears off eventually. But I would rather play
> a Nethack where there are fortune cookies/graffiti/imp taunts. Even
> when the novelty is gone, the cumulative effect of small amounts of
> flavour is large.

Nethack does a good job of generating flavour. It puts a lot of that
flavour text into the rare one in a hundred events rather than ensuring
every potion is generated inside a logical medicine cabinet.

> > > Roguelikes have the power of the written word at their disposal. Many
> > > of them throw it away in identikit settings and uninteresting messages.
> > > "You attack with your +0 long sword. You hit the giant bat. The giant
> > > bat dies." Yack.
> >
> > If they correctly conjugate the verb, I consider it a good start.
> > How do you generate your interesting messages if you cannot yet
> > generate the boring ones?
>
> Which verb did I conjugate incorrectly?

You conjugated them properly. That is the trick.

If one does: "%s hit %s", you need either a separate code path for
monsters versus the player, or automatically conjugate "hit". Since I
desire consistency, I go for the single code path so need to transform
hit to hits when it is the bat doing the hitting.

> > I tried something close to text adventures with You Only Live Once
> > (http://www.zincland.com/7drl/liveonce) It is close to a text
> > adventure in terms of having lots of prebuilt descriptions.
> > Replayability, however, suffers.
>
> I quite liked YOLO. Actually, I played it a couple of weeks ago, and I
> thought the text made it significantly more immersive and fun.
> Replayability suffers, but no 7DRL is going to be amazingly replayable,
> is it?

Thank you. I am glad you enjoyed it. I am very glad you read the text
too, as that took a fair amount of the development time :>

> (The other 7DRL I've played, the Sheep's Z-DAY, has virtually no
> textual description, and is very light on gameplay mechanics, but still
> has a lot of atmosphere, and I got a lot of enjoyment out of it for
> exactly that reason. "Zombie apocalypse" is one of my favourite
> settings.)

I think my major point here has been "do not model junk". Not everyone
agrees with me. I, however, feel if the only use of a chair is for
people to turn it into a improvised bludgeon, you shouldn't add it to
the game. Objects should be present due to *gameplay* uses, not
*reallife* uses.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Anonymous
June 29, 2005 6:28:18 PM

Archived from groups: rec.games.roguelike.development (More info?)

Jeff Lait <torespondisfutile@hotmail.com> schrieb:

> If one does: "%s hit %s", you need either a separate code path for
> monsters versus the player, or automatically conjugate "hit". Since I
> desire consistency, I go for the single code path so need to transform
> hit to hits when it is the bat doing the hitting.

Or you could just refer to the player by their name, thus eliminating
the need for using second person.

Also, some people need to snip their posts more...

--
Jim Strathmeyer
Anonymous
June 29, 2005 7:43:27 PM

Archived from groups: rec.games.roguelike.development (More info?)

In article <871x6lftox.fsf@dinopsis.dur.ac.uk>, Chris Morris <c.i.morris@durham.ac.uk> wrote:
>It wouldn't be suited to traditional ASCII. It'd obviously be okay
>with tiles, but it might also work with the zoomed ASCII suggested
>previously on the group.
>+--+--+--+
>|.@|**|.o|
>|..|!*|..|
>+--+--+--+
>
>Did anyone ever try zoomed ASCII? I've no idea how well it would work
>in practice. Possibly it'd be best just to go to tiles at that stage.

The really old DND game had a kind of zoomed ascii.

Alan
Anonymous
June 29, 2005 8:35:30 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Jeff Lait" <torespondisfutile@hotmail.com> writes:
> Chris Morris wrote:
> > "Jeff Lait" <torespondisfutile@hotmail.com> writes:
> > > The reason is that it is hard enough to make an unthemed dungeon. It
> > > is hard enough making a roguelike with items scattered on the floor
> > > without having to worry about ensuring each item has a *reason* to be
> > > there.
> >
> > *shrug* It doesn't have to be that much harder.
>
> I'm just bitter because after a few years I still haven't got to this
> "not much harder" part of roguelike development.

Perhaps it's just the way I implemented the creature/item chances in
mine.

Each level had a theme code. Each theme code had a list of items and
(relative) chances [1], and a list of creatures and chances. I started
out doing it this way back when I just had a single dungeon level
(with no particular theme). When I extended that to 9 levels of
generic caves (again with no particular theme) it still worked. When I
added the rat lair (only monsters generated are rats, only items
generated are things rats can't devour given sufficient off-screen
time [2]) it was very easy.

After that adding other themed levels (or just varying the mix of
items and creatures on a particular level) was straightforward because
I'd (purely through luck) written the item/creature generators so that
it *was* straightforward.

The big (and in retrospect obvious) disadvantage with this way of
doing things is that it took far too long to make the list for each
level and it was far too easy to miss things out (why are there
trousers on caverns 2 and 4, but not caverns 3?). I made a start on
fixes [3] to that but decided the system needed a bit more of a
rewrite to get it maintainable.

However, the *hard* bit in this was getting the item and creature
generation working at all. Once that was done making themed levels out
of it was easy. So that's what I meant by "not much harder" - the hard
bit was already done by the time I'd got one unthemed level;
everything else after that was data entry.

[1] So a short item chance file might be
SmallStick 3
BigStick 1
MetalBar 2
Clothes 2
ToughClothes 1
....and then it rolls a d9 to pick the item, 1-3 is the small stick, etc.

[2] Why do the rats have a knife lying on the floor? It was on the
adventurer, but they've already eaten that bit. That's as much of a
reason as it needs.

[3] The first big one was that, rather than having
HamSandwich 1
BeefSandwich 1
CheeseSandwich 1
it had
Sandwich 3
and when it generated a sandwich it just picked a sandwich type at random.

It really needed more than that doing, however. I currently have half
an outline of the 'right' system in my head, but transferring that to
data types needs some thought.

--
Chris
Anonymous
June 29, 2005 8:58:01 PM

Archived from groups: rec.games.roguelike.development (More info?)

At 28 Jun 2005 14:28:14 +0100,
Chris Morris wrote:

> "Jeff Lait" <torespondisfutile@hotmail.com> writes:
>> Timothy Pruett wrote:
> Well, if the items aren't on the ground because they're on the shelf,
> or the table, or whatever, it's still 'lying around' but not where
> monsters can trip over it. Similarly the healing potions might be in
> the locked medicine cabinet, which you could smash open, lockpick, or
> steal the key from the guard.

Just assume the furniture is not displayed for simplicity and clear
display reasons.

--
Radomir `The Sheep' Dopieralski @**@_
(Qq) 3 Sob?
. . . ..v.vVvVVvVvv.v.. .
Anonymous
June 29, 2005 9:31:38 PM

Archived from groups: rec.games.roguelike.development (More info?)

Quoting Matthew Bennett <mccb3@hotmail.com>:
>Maybe. But Nethack doesn't do this very consistently. It has grafitti
>("Vlad was here") that isn't important.

Indicates the presence of a trapdoor.

>important). Fortune cookies have messages in them, which aren't
>important.

About half contain useful hints.
--
David Damerell <damerell@chiark.greenend.org.uk> flcl?
Today is First Thursday, Presuary.
July 1, 2005 1:22:01 AM

Archived from groups: rec.games.roguelike.development (More info?)

Ray Dillinger wrote:
> Jeff Lait wrote:
>
> > I think my major point here has been "do not model junk". Not everyone
> > agrees with me. I, however, feel if the only use of a chair is for
> > people to turn it into a improvised bludgeon, you shouldn't add it to
> > the game. Objects should be present due to *gameplay* uses, not
> > *reallife* uses.
>
> I like junk. But I think care should be taken in its
> generation to make sure it appears in appropriate
> contexts.
>
> Several reasons:
>
> 1) Immersiveness. Dungeons aren't spotless places; if
> there isn't any junk around it feels weird. I'm
> not ready to believe a bunch of goblins in a bar
> with no drinking mugs, no barrels of horrible
> rotgut liquor, no chairs, no tables, and no
> tasteless pictures of goblin females (or males,
> if it's that kind of a bar) up on the wall. these
> elements distinguish "a bar" from "a pack of goblins
> in a room," and as such add variety to the experience
> of the game.

I agree on this one. I place some props in goblin encounters ("a small
smoky fire", "a pile of toasted rats", etc). Similarly in wizard
encounters ("a tall ebony chair", "a quill pen", etc). And so on.

> 2) Clues. Lairs of large carnivores frequently have
> gnawed bones around them; intelligent monsters create
> middens; slime trails usually indicate slugs or
> gelatinous somethingorothers; flint shards often
> indicate the presence of people who make and use
> stone weapons; etc. I think junk should indicate
> what kind of monsters are likely to be nearby, and
> that makes most junk useful. (Since most RL's don't
> do this, this doesn't apply to, EG, angband junk).

Yes. Though in Guild you don't tend to start finding useful item clues
until you actually go into the room where the monster is. Long-distance
warnings are done by sound instead ("You hear thunderous snoring").

The line between junk and good items blurs. I decided in Guild that I
wanted a monster that uses equipment to drop equipment when it died -
for more flavor and realism. So an animated skeleton always drops 'a
human skeleton', 'a shield' and 'a longsword'. Although these are
potentially useful for a low level character, they are junk when the
party gets more experienced. Adds flavor but can be pesky until you get
used to ignoring drops of weak monsters.

Another use of junk is to camouflage good stuff. Like in a carrion
crawler lair in Guild where you have to rummage through the corpses
and body parts to get any treasures that may be there. It gives you a
'treasure hunt' feeling that wouldn't be there if the items were just
lying on the floor in plain view.

A.
July 1, 2005 1:38:55 AM

Archived from groups: rec.games.roguelike.development (More info?)

Chris Morris wrote:
> "Shedletsky" <mylastname@stanford.edu> writes:
> > > 3) I want monumental battles. I want to play my small part in some
> ...
> > > to see a game do this though. Nothing would be better than being able to
> > > take my level 43 Ranger and have him shoot down orc after orc from the top
> > > of the keep, alongside dozens of fellow archers.
> >
> > I like this idea. Are there any existing RLs in which you can hire/acquire
> > computer controlled NPCs?
>
> ADOM, Nethack, Crawl (all definitely acquire rather than hire, with
> one partial exception from ADOM that I know of)
>
> Oh, and my roguelike had bits of this, and was going to have more (but
> that's on hold while I rewrite the engine to be fast enough to cope) -
> you could rescue a merchant and his guards from a horde of rats, for example
> (though I never quite got that encounter balanced properly, since the
> guards were perfectly capable of taking on the rats themselves...)
> Not quite an epic battle, but as much of one as would fit on the maps
> I was using.

Guild has 'battle-between-some-orcs-and-some-goblins' encounters. They
are quite cool, but have some downsides. Usually the encounter is over
by the time you get there, with the goblins either inflicting some
casualties and running away, or getting trapped in a dead end and
slaughtered. (Still, you can hear the sounds of battle, see the corpses
when you reach the scene, and perhaps mop up some of the retreating
goblins...)

Also, they really slow the game down due to the amount of extra AI
computations that occur each turn, so I kept them to less than 10
individuals on each side (and even then the lag is perceptible).

So, I'd like to insert more monsters-vs-monsters battle encounters in
future releases, but nothing epic as it would slow the game down too
much.

Come to think of it, it would be neat to put in an encounter where an
NPC is found fighting against monsters and will join your group if
he/she survives. I know this sounds like vaporware, but actually the
program is just about there already - I reckon I could implement this
in about 3 hours of coding. (I think I'd have to place it in a themed
level to make sure the NPC-monsters fight appears somewhere close to
the stairs so you can intervene before it's over...)

A.
Anonymous
July 1, 2005 4:38:44 AM

Archived from groups: rec.games.roguelike.development (More info?)

Jeff Lait wrote:

> I think my major point here has been "do not model junk". Not everyone
> agrees with me. I, however, feel if the only use of a chair is for
> people to turn it into a improvised bludgeon, you shouldn't add it to
> the game. Objects should be present due to *gameplay* uses, not
> *reallife* uses.

I like junk. But I think care should be taken in its
generation to make sure it appears in appropriate
contexts.

Several reasons:

1) Immersiveness. Dungeons aren't spotless places; if
there isn't any junk around it feels weird. I'm
not ready to believe a bunch of goblins in a bar
with no drinking mugs, no barrels of horrible
rotgut liquor, no chairs, no tables, and no
tasteless pictures of goblin females (or males,
if it's that kind of a bar) up on the wall. these
elements distinguish "a bar" from "a pack of goblins
in a room," and as such add variety to the experience
of the game.

2) Clues. Lairs of large carnivores frequently have
gnawed bones around them; intelligent monsters create
middens; slime trails usually indicate slugs or
gelatinous somethingorothers; flint shards often
indicate the presence of people who make and use
stone weapons; etc. I think junk should indicate
what kind of monsters are likely to be nearby, and
that makes most junk useful. (Since most RL's don't
do this, this doesn't apply to, EG, angband junk).

3) Imagination. Junk allows you to use your players'
imaginations to come up with ideas you can use to
deepen the game. You may not think chairs have any
use in your dungeon, but within a month of making
them exist, you'd get three bug reports. The first:
"I tried jamming a door with a chair, and it didn't
work." The second: "I needed to make a fire, but
the chairs don't chop apart into firewood." and the
third: "If you stand on a chair, you ought to be
able to reach things on the cieling." And these
are all good ideas that add flexibility, immersiveness,
and realism to the game without breaking it by making
something uber-powerful.
Anonymous
July 1, 2005 5:25:48 PM

Archived from groups: rec.games.roguelike.development (More info?)

Quoting Matthew Bennett <mccb3@hotmail.com>:
>Jeff Lait wrote:
>>Modelling junk is always a concern for me. I'm very much in the
>>Nethack school of thought: The dungeon *has* bones, medicine cabinets,
>>whatever, they just aren't shown if they aren't important.
>Maybe. But Nethack doesn't do this very consistently.

I completely missed the point here; specifically what NH avoids modelling
are junk _objects_. There are a very few exceptions, but almost every
object has some use.
--
David Damerell <damerell@chiark.greenend.org.uk> Kill the tomato!
Today is First Saturday, Presuary - a weekend.
Anonymous
July 6, 2005 7:51:54 PM

Archived from groups: rec.games.roguelike.development (More info?)

"Timothy Pruett" <drakalor.tourist@gmail.com> wrote in message
news:E-idnTBTTad-913fRVn-rA@adelphia.com...
> 3) I want monumental battles. I want to play my small part in some
> epic battle at the gates of Mordor, or defending the people of Rohan
> in Helm's Deep. I want to fight alonside dozens of fellow soldiers,
> against a sea of enemies. I want seige warfare, catapults flinging
> flaming debris and giant boulders, battering rams pounding on the
> gate, soldiers digging a tunnel beneath a section of castle wall in
> order to collapse it, and mighty heroes and villains clashing in the
> midst of a great battle. I want all of it. But, it probably won't
> happen. Such a game would need to make large-scale combat as easy to
> manage as small-scale one-on-one fighting in the dungeons. It's
> doable, but quite the challenge. I'd love to see a game do this
> though. Nothing would be better than being able to take my level 43
> Ranger and have him shoot down orc after orc from the top of the keep,
> alongside dozens of fellow archers.

Have you played Dynasty Warriors (3, 4 or 5)? I enjoy those games
quite a bit (especially the 2-player mode), mainly due to the feeling of
fighting in epic battles.

I did toy with a DW-esque engine some time ago, but could never really
make it work in a roguelike. Still, if you have the means, I suggest
them as reference material.

--
Glen
L:p yt E+++ T-- R+ P+++ D+ G+ F:*band !RL RLA-
W:AF Q+++ AI++ GFX++ SFX-- RN++++ PO--- !Hp Re-- S+
Anonymous
July 27, 2005 12:53:34 PM

Archived from groups: rec.games.roguelike.development (More info?)

On 30 Jun 2005 21:22:01 -0700
"Antoine" <mail@guildgame.com> wrote:

>Adds flavor but can be pesky until you get
>used to ignoring drops of weak monsters.

It doesn't matter how much junk you have, as long as you have a wand of polymorph/change.
Anonymous
August 3, 2005 1:42:14 PM

Archived from groups: rec.games.roguelike.development (More info?)

So far, I like a lot of Guild's dungeon "junk", as well as some other
nice factors in that game. Really superb RL!

Really though, tautology and deus ex machina seem to be big parts of
Roguelike games. The dungeon exists because it exists, and the
minotaurs are there to give your character enough preparation/loot to
face the evil archlich.

The fact that I am branching away from the traditional RL in my
(currently vaporware - but I downloaded pdcurses last night!) game is
hopefully not going to be viewed as a stigma. I would like to have an
interesting story and background, and possibly implement slightly
amusing ASCII cutscenes following plot-important quests. Im not quite
sure how much this is going to effect the flavor of the game though.
It's hard to predict if this would work with everything a conventional
Roguelike game can do. Should I implement wands of polymorphism...
digging through dungeons... there's a lot of stuff I'm still unclear
on, and probably will remain very unclear on until I at least get a
decent working beta up and running. All I know how is: random dungeons,
ascii art, some work on enemy AI, and a lot of random stuff to find.
August 3, 2005 5:58:32 PM

Archived from groups: rec.games.roguelike.development (More info?)

Wildhalcyon wrote:
> So far, I like a lot of Guild's dungeon "junk", as well as some other
> nice factors in that game. Really superb RL!

That's what we like to hear :) 

How far through have you got?

A.
Anonymous
August 4, 2005 2:24:32 PM

Archived from groups: rec.games.roguelike.development (More info?)

Only to level 4 in the small cave (my characters are still lvl 1), but
the house had some interesting belongings too like chests and beds and
wardrobes. In the dungeon, here are a couple chests scattered about,
but most of the stuff I've found has been laying about. I don't mind
having barrels and chests and such to have to open to get the contents.
One more button push isnt going to kill me. Adding some creative/random
adjectives could personalize it more "There is an oak chest"/"There is
a rusted chest" etc. perhaps adding various modifiers to damage dealt.
I think most objects in the game should be treated as inanimate
monsters - give them defense and HP. Normal chests might take a couple
swift kicks to eliminate, but a sturdier Ironwood chest might take 10,
and a magically enchanted chest might retaliate by casting a freeze
spell.. etc.

In Arcanum (which I'm using to base some of my skills off of - its not
really a Roguelike, but its loads of fun), you can "attack" inanimate
objects by holding ALT and clicking. holding the ALT might not work
with a DOS RL - and using the mouse isnt really an option, but it seems
like it might be a more reasonable approach might be to hold the shift
key with a direction key?
!