Skills

1 view
Skip to first unread message

Archibald

unread,
Aug 15, 2001, 4:25:08 PM8/15/01
to
What skills we can use in RL game? If you know more, add it to the list.


ATTRIBUTE BONUS / FIGHT:
Weapon bonus - (sword, axe, bow)
Monster bonus - (dragon slayer, exorcist)
Shield bonus
Armour - ignore bad effects/allow heavy armour
Body Building - HP bonus
Athlectics - Physic attr bonus (STR, DEX)
Morale - ignore fear, panic
Critical hit

MAGIC:
Spell power bonus (fire, cold, energy, teleportation, enchant, death)
Concentration - Meditation - MP bonus
Lower casting cost (MP)

MOVEMENT (World):
Use transport - (Horse riding, boat)
Portal control - allow use world portals
Access area - Climbing, Swimming, Pathfinding
Cartography - Map making - allow map?
Bridge building

MOVEMENT (Dungeon):
Searching
Detect traps
Stealth
Pick lock (doors)
Teleport self
Phasing walls (ignore walls)
Swimming
Direction sense (?)
Danger sense

TREASURE:
more artifacts/rare items
more gold from monsters (more furs, teeth, etc. for sale)
more food (survival, hunting)
Haggling
Pick pocket
Appraising
Identify item

OTHERS:
Healing - fast healing rate
Health - poison, bleeding less deadly
Repair item
Creating weapons/armours (Smithing)
Creating potions (alchemy)
Enchanting - adding magic bonus for weapons etc.
Pet control - max pet number
Leadership - bonus for ally
Learning (exp bonus)

Chris Subich

unread,
Aug 16, 2001, 12:04:15 AM8/16/01
to
On Wed, 15 Aug 2001 22:25:08 +0200, Archibald <kko...@kki.net.pl>, the
evil little half-monkey he is, wrote:

>What skills we can use in RL game? If you know more, add it to the list.

What kind of Roguelike are you going for? Will you use a skill-tree
method?

>
>
>ATTRIBUTE BONUS / FIGHT:
>Weapon bonus - (sword, axe, bow)
>Monster bonus - (dragon slayer, exorcist)
>Shield bonus
>Armour - ignore bad effects/allow heavy armour
>Body Building - HP bonus
>Athlectics - Physic attr bonus (STR, DEX)
>Morale - ignore fear, panic
>Critical hit

Morale is not really a skill in and of itself. Likewise, critical hits
represent exceptional strikes with your weapon -- they're not usually aimed
for.

An athletics skill, by providing statistic boni, conflicts with separation
of skills and statistics. Statistics, as fundamental character attributes,
should provide boni to skills, and not the other way around.

Add parry and dodge.

>
>MAGIC:
>Spell power bonus (fire, cold, energy, teleportation, enchant, death)
>Concentration - Meditation - MP bonus
>Lower casting cost (MP)

For concentration, see the above.

I'd recommend grouping spells under general genres, like evocation,
divination, foo elemental, etc., and inheriting experience from individual
spells to the genre which provide a bonus to all spells in the genre.

Lower casting costs would go with individual spell proficency.

>
>MOVEMENT (World):
>Use transport - (Horse riding, boat)
>Portal control - allow use world portals
>Access area - Climbing, Swimming, Pathfinding
>Cartography - Map making - allow map?
>Bridge building

Mount, bridge building, and the Access Area set would probably all go under
woodcraft.

"Allow use [of] world portals?" They're static objects, not living things
-- if you can throw a bowling ball in and have it come out on the other
side, a person should be able to use 'em without a skill check.

Cartogrophy is a necessary skill -- if the PC's don't have it, the players
will start making maps on their own, and it will be frustratingly
mind-numbing for them.

>
>MOVEMENT (Dungeon):
>Searching
>Detect traps
>Stealth
>Pick lock (doors)
>Teleport self
>Phasing walls (ignore walls)
>Swimming
>Direction sense (?)
>Danger sense

Phase walls and teleport self are both spells. A case could be made for
both direction and danger senses as either spells or woodcraft, although
why you'd need direction sense in a roguelike is beyond me.

Searching [detect traps and secret doors], hide, move silently, and pick
lock are all thiefy skills.

>
>TREASURE:
>more artifacts/rare items
>more gold from monsters (more furs, teeth, etc. for sale)
>more food (survival, hunting)
>Haggling
>Pick pocket
>Appraising
>Identify item

The first three aren't skills. There is no skill that you could possibly
ever, ever use to actually create items or gold that wasn't there before
(or wouldn't be for someone else). If, by "more furs" you mean being able
to skin more and better furs, that's skinning and is a woodcraft skill.

Pick pocket is a thiefy skill. Identify item is either a spell or an
appraisal (as an appraisal is just a limited identification), your pick.

>
>OTHERS:
>Healing - fast healing rate
>Health - poison, bleeding less deadly
>Repair item
>Creating weapons/armours (Smithing)
>Creating potions (alchemy)
>Enchanting - adding magic bonus for weapons etc.
>Pet control - max pet number
>Leadership - bonus for ally
>Learning (exp bonus)

Healing is a valid skill, and I would say it goes under woodcraft because
you use local herbs & the like as tools.

Health is a statistic called Constitution. See above on athletics and
concentration.

Repair item, smith mundane item are both smithy skills.

Enchanting and alchemy both seem to belong primarially under magic.

Pet control and leadership are both things you get from Charisma, so no
skill for them.

Learning is intelligence, not a skill.
--
Chris # csubich@g-d-i-.-n-e-t
# (remove dashes to email)
"If by 'working perfectly', you mean 'totally ****** up',
that would be correct." - Ron Hiler

Björn Bergström

unread,
Aug 16, 2001, 7:21:12 AM8/16/01
to

"Archibald" <kko...@kki.net.pl> skrev i meddelandet
news:MPG.15e4f5f58...@news.tpi.pl...

I have this non-complete list of skills among my dev.notes. Please note
that this list is based on a post by Michael Hahn dated 2001-07-28
23:30:13 PST
(http://groups.google.com/groups?q=skill+tree&hl=sv&group=rec.games.rogu
elike.development&safe=off&rnum=4&selm=V3O87.420%247d.93632%40newshog.ne
wsread.com). Many skills are RPG skills that might not be very useful in
a roguelike (dancing, cooking etc). Use at will...

Combat
Melee
Offense
Blades
Defense
Disarm - Make the opponent drop his weapon

More skills are needed in the combatbranch. Together with blades you
should have axes, polearms etc and then ofcourse a branch called
Ranged...

Craft
Armourer - Create armour from raw materials
Leather
Metal
Carpentry
Boatbuilding
Housebuilding
Cooking
Dancing
Jewelry
Musical Instrument
Singing
Trade - Shoemaking, researching
Weaponsmithing - Create weapons from raw materials
Blades
Axes
Polearms
Bows
Crossbows


Survival
Climbing - Climping without a rope or ladder
Direction Sense - Tell which way is North
Hunting
Find/Remove Trap
Set Trap
Riding
Survival
Forest
Desert
Mountain
Swim
Tracking
Weather Sense


Magical
Arcana - Dynamic magic skills
Sorcery - Spellcasting
Targeting - Directing spells to their targets
Sixth Sense - Sensing magical energies
Summoning
Elemental
Planer
Enchanting
Static - Major, permanent enchantments
Temporary - Scrolls and potions

For tips on magicsystems have a look at my dev section at
home.swipnet.se/dungeondweller/index.htm click on roguelike development.

Knowledge
Alchemy - Brew potions
Appraising - Estimate value of object
Herbalism - Identify plants, brew mixtures and poisons
History
Language
Literacy
Magical Theory - Research and experimentation
Mathematics
Medicine
First Aid
Healing
Surgery
Religion
Zoology


Thievery
Acrobatics
Disguise
Forgery
Hide
Jumping - Extraodinary leaps
Pick Pocket
Sneak - Ability to move with stealth

All the best,
Björn Bergström - dungeon...@swipnet.se
Dungeondweller RLG and Roguelike Development Articles at
http://home.swipnet.se/dungeondweller/index.htm


Michael Blackney

unread,
Aug 17, 2001, 2:24:31 AM8/17/01
to
pfh...@gdi.net (Chris Subich) wrote...

> On Wed, 15 Aug 2001 22:25:08 +0200, Archibald <kko...@kki.net.pl>, the
> evil little half-monkey he is, wrote:
>
> >What skills we can use in RL game? If you know more, add it to the list.
>
> What kind of Roguelike are you going for? Will you use a skill-tree
> method?
>
> >
> >
> >ATTRIBUTE BONUS / FIGHT:
> >Weapon bonus - (sword, axe, bow)
> >Monster bonus - (dragon slayer, exorcist)
> >Shield bonus
> >Armour - ignore bad effects/allow heavy armour
> >Body Building - HP bonus
> >Athlectics - Physic attr bonus (STR, DEX)
> >Morale - ignore fear, panic
> >Critical hit
>
> Morale is not really a skill in and of itself. Likewise, critical hits
> represent exceptional strikes with your weapon -- they're not usually aimed
> for.

A skill such as Monster Lore should give benefits to morale tests such
as fear and panic. If you have a system which allows feats (as in
d20) or intrinsics (like in Nethack) you could include a Nerves of
Steel (or some such) ability to represent characters of particular
fortitude.
I don't mind the Find Weakness skill in ADOM. Critical hits often
represent well placed blows - an Anatomy skill could help here, along
with the above mentioned Monster Lore.

>
> An athletics skill, by providing statistic boni, conflicts with separation
> of skills and statistics. Statistics, as fundamental character attributes,
> should provide boni to skills, and not the other way around.
>

I agree. Engaging in athletic activity should exercise physical
attributes as well as training athletic skills (such as Jump,
Swimming, Sprint). Only characters who are created with high
athletics skill should recieve a bonus to physical attributes (once
off, if at all).

> Add parry and dodge.

It depends on your level of abstraction and implementation. I am not
a huge fan of skills like sword, longsword, etc. When you are
fighting it matters less what you are using than it does how and where
you are using it. For the how, I would add: parry, dodge, feint,
riposte, lunge, accept charge, drive and retreat. These skills would
have bonuses for different weapons (ie a sword, shield or staff would
give a parry bonus where a polearm would give an accept charge bonus).
For the where, I would add: field combat, enclosed combat, blind
combat, intoxicated combat, and possibly a skill for combat on
non-level ground. (Feel free to offer better names for these.
Please!) Again, some weapons would benefit from being used in
different locations and under different conditions. A dagger would
give an enclosed combat bonus, long weapons would benefit from field
combat, weapons which hit easily (ie. flails & shields) would work
well in the dark.

>
> >
> >MAGIC:
> >Spell power bonus (fire, cold, energy, teleportation, enchant, death)
> >Concentration - Meditation - MP bonus
> >Lower casting cost (MP)
>
> For concentration, see the above.

I like having an attribute for concentration. It is useful for magic
users (casting) and thieves (hiding, moving silently) but also helps
whenever the PC wants to do more than one thing at a time (running and
reading a scroll, for instance).

>
> I'd recommend grouping spells under general genres, like evocation,
> divination, foo elemental, etc., and inheriting experience from individual
> spells to the genre which provide a bonus to all spells in the genre.
>
> Lower casting costs would go with individual spell proficency.
>
> >
> >MOVEMENT (World):
> >Use transport - (Horse riding, boat)
> >Portal control - allow use world portals
> >Access area - Climbing, Swimming, Pathfinding
> >Cartography - Map making - allow map?
> >Bridge building
>
> Mount, bridge building, and the Access Area set would probably all go under
> woodcraft.
>
> "Allow use [of] world portals?" They're static objects, not living things
> -- if you can throw a bowling ball in and have it come out on the other
> side, a person should be able to use 'em without a skill check.
>

True. Though it does depend on your world. I don't like the idea of
regular static portals, so I'd include portal use under some catch-all
magic knowledge skill.

> Cartogrophy is a necessary skill -- if the PC's don't have it, the players
> will start making maps on their own, and it will be frustratingly
> mind-numbing for them.
> >
> >MOVEMENT (Dungeon):
> >Searching
> >Detect traps
> >Stealth
> >Pick lock (doors)
> >Teleport self
> >Phasing walls (ignore walls)
> >Swimming
> >Direction sense (?)
> >Danger sense
>
> Phase walls and teleport self are both spells. A case could be made for
> both direction and danger senses as either spells or woodcraft, although
> why you'd need direction sense in a roguelike is beyond me.

I have a wilderness

>
> Searching [detect traps and secret doors], hide, move silently, and pick
> lock are all thiefy skills.
>
> >
> >TREASURE:
> >more artifacts/rare items
> >more gold from monsters (more furs, teeth, etc. for sale)
> >more food (survival, hunting)
> >Haggling
> >Pick pocket
> >Appraising
> >Identify item
>
> The first three aren't skills. There is no skill that you could possibly
> ever, ever use to actually create items or gold that wasn't there before
> (or wouldn't be for someone else).

Perception should be a skill, and might allow marginally better
corpse-searching ability (though generally just for small things like
gold, tic-tacs, etc.)

> If, by "more furs" you mean being able to skin more and better furs, that's
> skinning and is a woodcraft skill.

Unless you are skinning animals not made of wood. Then it's a hunting
skill. ;)

>
> Pick pocket is a thiefy skill. Identify item is either a spell or an
> appraisal (as an appraisal is just a limited identification), your pick.
>
> >
> >OTHERS:
> >Healing - fast healing rate
> >Health - poison, bleeding less deadly
> >Repair item
> >Creating weapons/armours (Smithing)
> >Creating potions (alchemy)
> >Enchanting - adding magic bonus for weapons etc.
> >Pet control - max pet number
> >Leadership - bonus for ally
> >Learning (exp bonus)
>
> Healing is a valid skill, and I would say it goes under woodcraft because
> you use local herbs & the like as tools.
>
> Health is a statistic called Constitution. See above on athletics and
> concentration.

Healing ability might be a skill, though I'd prefer not.

> Repair item, smith mundane item are both smithy skills.
> Enchanting and alchemy both seem to belong primarially under magic.
>
> Pet control and leadership are both things you get from Charisma, so no
> skill for them.

You don't think that taking control of animals and people takes
practice? I might go so far as to say that charisma might be able to
be removed completely from the stats and replaced by skills (yes, some
people can do this naturally, but some people can naturally jump
higher and play the violin).

> Learning is intelligence, not a skill.

Very true.

Your list looks quite good. Develop it more and bring it back, I'd
love to see it.

Michael Blackney

R Dan Henry

unread,
Aug 19, 2001, 8:30:24 PM8/19/01
to
On Thu, 16 Aug 2001 04:04:15 GMT, the disembodied brain of
pfh...@gdi.net (Chris Subich) transmitted thus:

>On Wed, 15 Aug 2001 22:25:08 +0200, Archibald <kko...@kki.net.pl>, the
>evil little half-monkey he is, wrote:
>
>>What skills we can use in RL game? If you know more, add it to the list.

Anything you can think of a use for.

>>ATTRIBUTE BONUS / FIGHT:
[snip]


>>Body Building - HP bonus
>>Athlectics - Physic attr bonus (STR, DEX)
>>Morale - ignore fear, panic
>>Critical hit

>Morale is not really a skill in and of itself.

There's no reason it cannot be treated as one, although I'd call it
something like Nerve or Poise myself.

>Likewise, critical hits
>represent exceptional strikes with your weapon -- they're not usually aimed
>for.

On the contrary, having a skill to improve critical hit chances allows
you to have high-skilled but wild-swinging Barbarian-Dervishes and
high-skilled, precise Expert Duelists in your world. The former would
develop the ability to hit consistently and hit hard, whereas the
later would develop the ability to hit consistently and to find the
weaknesses in an opponents defenses. The difference between 2d6+4
every turn with only lucky crits and 2d6 plain, but frequent crits.
This makes more sense if crits don't just add damage, but have other
effects, like bypassing armor or bypassing random hit location to hit
a vital area. Precision is a good name for this skill and if it gets
big bonuses on unaware targets, it is Assassination, too.

>An athletics skill, by providing statistic boni, conflicts with separation
>of skills and statistics.

Better complain about body-building, too, then, it's taking over CON.

>Statistics, as fundamental character attributes,
>should provide boni to skills, and not the other way around.

Why? Sure, this is the way it's normally been done, but it doesn't
have to be. You really don't even need to have stats in skill-based
game. A strong character is one with high brute force skills, a
dexterous character is one with lots of acrobatic skills, a
charismatic character is one with lots of social skills. It isn't as
if stats are pristine and sacred in roguelikes, anyway. Most, if not
all, developed roguelikes involve a process of maxing out your stats,
same as developing skills. In Nethack you can even do some of this by
practicing the stat, just like practicing a skill to improve. In real
life, practicing relevant skills boosts the stat.

Stats are just an abstraction, much more so than skills. You get
stronger by lifting weights or rowing or other exercise, but which
muscles are strong in what proportions depends on how you have
exercised. If you do lots of word games and crossword puzzles, you'll
do wonders for your verbal intelligence, but if you never juggle any
numbers, your mathematical intelligence will rot. You can learn to
juggle or you can learn to do slight-of-hand, but an aptitude for one
use of 'dexterity' does not imply an aptitude for the other. I think
feedback to stats from skills is a good step in the direction of
realism, if that's what you want. Elimination might even be better,
stats were a crude generalization of abilities created for class/level
based gaming. A good skill set can define a character better. Some
'aptitudes' makes sense. A character with the 'Bookish' aptitude would
do better at academic skills, without implying that this means greater
intelligence in general.

>Add parry and dodge.

Oh, yeah. Defensive skills are a must. Once you have a dodge skill,
that's about 50% of the old DEX stat taken care of, right there.

>>MAGIC:
>>Spell power bonus (fire, cold, energy, teleportation, enchant, death)
>>Concentration - Meditation - MP bonus
>>Lower casting cost (MP)

>For concentration, see the above.

??? You want magic points to be purely stat based? If so, I disagree.
Skill is important. Level is at least as important as spell stat in
your standard roguelike. To truly do a skill-based game means ditching
levels, which means skill has to do the work that level did. You have
to buy those magic point advances now by developing the appropriate
skill(s).

>I'd recommend grouping spells under general genres, like evocation,
>divination, foo elemental, etc., and inheriting experience from individual
>spells to the genre which provide a bonus to all spells in the genre.

Depends on how many spells you have. Not worth it if you only have 50
spells. And your approach creates complication in multiple type
spells. Is summon fire elemental to use 'fire magic' skill or
'summoning' skill? Or the game designer may want to have character
race determine base skill in a magical type, and only allow skill to
enhance particular skills.

You really have total freedom in designing your magic system, as there
aren't any realism issues.

>Lower casting costs would go with individual spell proficency.

I agree on this point. Improved general efficiency is better modeled
as increased spell points.

>>MOVEMENT (World):
>>Use transport - (Horse riding, boat)
>>Portal control - allow use world portals
>>Access area - Climbing, Swimming, Pathfinding
>>Cartography - Map making - allow map?
>>Bridge building

A skill to double movement speed on a large-scale map only? Distance
Running? Or that could be your Pathfinding?

>Mount, bridge building, and the Access Area set would probably all go under
>woodcraft.

Woodcraft? Climbing & swimming are skills for the urban thief. Riding
for the knight who never found his own food in his life. Bridge
building is best done by a team of engineers with sound educations. A
woodsman may be able to drop a log across a stream, but that's about
it. For rivers, even small ones, you need to know what you are doing,
and that isn't part of woodcraft. And horses are creatures of the open
plain. The more wooded the area, the less useful they become.

>"Allow use [of] world portals?" They're static objects, not living things
>-- if you can throw a bowling ball in and have it come out on the other
>side, a person should be able to use 'em without a skill check.

On the other hand, maybe you could get lost if you don't know what you
are doing. Or die. He obviously has something magical in mind, in
which case it can work however he pleases.

>Cartogrophy is a necessary skill -- if the PC's don't have it, the players
>will start making maps on their own, and it will be frustratingly
>mind-numbing for them.

I wholeheartedly agree. Roguelikes require automapping. On the other
hand, maybe there's some other use for such a skill.

>>MOVEMENT (Dungeon):
>>Searching
>>Detect traps

These really aren't movement skills. These are observation spells.

>>Stealth
>>Pick lock (doors)

If you can pick a lock on a door, I'll bet you can pick a lock on a
chest, too. :-) Wouldn't call this movement, myself.

>>Teleport self
>>Phasing walls (ignore walls)
>>Swimming
>>Direction sense (?)

Like mapping, this is pretty much got to be an automatic. We can see
which way is toward the top of the monitor and nothing else should
matter. This would be an observation skill anyway, not movement.

>>Danger sense

Observation, not movement.

>Phase walls and teleport self are both spells.

For *humans* phasing through walls and teleporting are spell effects.
For quantum pixies, they may be skills.

>Searching [detect traps and secret doors], hide, move silently, and pick
>lock are all thiefy skills.

I disagree with the stereotyping. Hiding, for example, is a good skill
for anyone. And if you have a woodcraft skill area, moving silently in
the wilderness would go there.

>>TREASURE:
>>more artifacts/rare items

Crafting/repair abilities cover this.

>>more gold from monsters (more furs, teeth, etc. for sale)

Butchering. Plus some form of Appraisal to know what to save.

>>more food (survival, hunting)

Cooking, too.

>>Haggling
>>Pick pocket
>>Appraising
>>Identify item

>The first three aren't skills. There is no skill that you could possibly
>ever, ever use to actually create items or gold that wasn't there before
>(or wouldn't be for someone else).

No, but I don't think he means creating them from nothing. (Although
some systems use 'luck' to improve treasure found. Hey, it's a fantasy
game.) He's using very shorthand notation here, so it isn't always
clear. Craft skills, appraisal skills, observation skills, and
bargaining skills can all improve an adventurers ultimate haul. Not to
mention a possible Portage skill to increase the amount one can carry.

>If, by "more furs" you mean being able
>to skin more and better furs, that's skinning and is a woodcraft skill.

When you mean 'general wilderness-type skill', please don't call it
'woodcraft'. Much wilderness is not wooded.

>Pick pocket is a thiefy skill.

One of the reasons for going to skills system is to leave behind this
kind of automatic stereotyping and open up character development to
something more flexible than fighter/mage/priest-healer/thief
stereotypes.

>Identify item is either a spell or an
>appraisal (as an appraisal is just a limited identification), your pick.

Identifying and appraising are two different things. A bard would
sooner recognize the fabled Mandolin of the Highly Annoying Minstrel.
But he'd be less equipped than his burglar pal to appraise it, i.e. to
judge its market value.

>>OTHERS:
>>Healing - fast healing rate
>>Health - poison, bleeding less deadly
>>Repair item
>>Creating weapons/armours (Smithing)
>>Creating potions (alchemy)
>>Enchanting - adding magic bonus for weapons etc.
>>Pet control - max pet number
>>Leadership - bonus for ally
>>Learning (exp bonus)

>Healing is a valid skill, and I would say it goes under woodcraft because
>you use local herbs & the like as tools.

You are more likely to be using wood, cloth, and mud. Herbs are good
for a variety of conditions, but spurting blood isn't one of them. A
prudent fighter will learn the art of first aid whether he dwell in
the city or woodland village.

>Health is a statistic called Constitution. See above on athletics and
>concentration.

Again, there really is no need to keep stats if you have a good skill
system. You are much too willing to accept 'the way it has always been
done' as the only way.

>Enchanting and alchemy both seem to belong primarially under magic.

But they don't have to be treated as *spells*, so they are better
grouped with other crafts than with spell-casting unless they *are*
treated as spells. In many fantasy worlds and myths, creation of an
enchanted sword is the ultimate power of smithing. At a certain high
point of skill, a crafter becomes able to put a bit of himself into
his work, creating a 'magic' item.

>Pet control and leadership are both things you get from Charisma, so no
>skill for them.

Right. Nobody ever learned anything about handling animals or dealing
with people. You're born with it, or don't even bother. Sure.

>Learning is intelligence, not a skill.

I disagree with your phrasing (because you appear to be basing this on
stat-centered thinking), but I agree that this shouldn't be a skill. A
general experience boost from a skill is likely to be seriously
abusable. However, I could see a Studying skill that would improve
your ability to absorb training. Good study habits do matter. And a
high Studying skill could make it easier to find a teacher, as that
wizard may not want to pass on his arcane knowledge of the Making
Pointy Sticks Fly At Your Enemies spell if he thinks he might be
wasting his time.

--
R. Dan Henry, Emperor of the Universe
rdan...@earthlink.net
Redneck -- the other white meat

Chris Subich

unread,
Aug 19, 2001, 11:22:17 PM8/19/01
to
On Mon, 20 Aug 2001 00:30:24 GMT, R Dan Henry <rdan...@earthlink.net>,
the evil little half-monkey he is, wrote:

>On the contrary, having a skill to improve critical hit chances allows
>you to have high-skilled but wild-swinging Barbarian-Dervishes and
>high-skilled, precise Expert Duelists in your world. The former would
>develop the ability to hit consistently and hit hard, whereas the
>later would develop the ability to hit consistently and to find the
>weaknesses in an opponents defenses. The difference between 2d6+4
>every turn with only lucky crits and 2d6 plain, but frequent crits.
>This makes more sense if crits don't just add damage, but have other
>effects, like bypassing armor or bypassing random hit location to hit
>a vital area. Precision is a good name for this skill and if it gets
>big bonuses on unaware targets, it is Assassination, too.

An Expert Duelist, however, is not going to be toting a two-handed
broadsword, and a barbarian is not going to bt playing with the sharp
pointy stick that other people call rapiers.

D&D has critical range based upon weapon type, and I happen to aggree with
that -- perhaps allow existing combat boni to come into play [say the
weapon has a critical hit ratio, and you criticially hit when your roll +
boni is more than ratio times defender's roll + boni].


In my mind, skills should be something definite. In the case of combat,
it's something you do -- swing a broadsword or dodge. A critical hit, to
me, is a lucky or expert result from a hit -- you usually aren't aiming to
do a critical hit over and above aiming to hit and cause damage to your
enemy. A critical hit is a subset of a hit, so I don't think it needs its
seperate skill.


>
>>An athletics skill, by providing statistic boni, conflicts with separation
>>of skills and statistics.
>
>Better complain about body-building, too, then, it's taking over CON.

That was a listed skill? I missed it.

>
>>Statistics, as fundamental character attributes,
>>should provide boni to skills, and not the other way around.
>
>Why? Sure, this is the way it's normally been done, but it doesn't
>have to be. You really don't even need to have stats in skill-based
>game. A strong character is one with high brute force skills, a
>dexterous character is one with lots of acrobatic skills, a
>charismatic character is one with lots of social skills. It isn't as
>if stats are pristine and sacred in roguelikes, anyway. Most, if not
>all, developed roguelikes involve a process of maxing out your stats,
>same as developing skills. In Nethack you can even do some of this by
>practicing the stat, just like practicing a skill to improve. In real
>life, practicing relevant skills boosts the stat.

True on the last point, but that's a function of the skill making use of
the stat. Simply knowing athletics does not make me stronger, it's the use
of it (which also makes use of the stats involved) that makes me stronger.

I really happen to like the skill tree idea -- stats up at the very top,
and skills make use of stats to modify their rolls, but the simple presence
or absence of a skill does not change the stat. This also prevents
unwanted codependency between independent stats, e.g. where your high skill
score im barbell lifting makes you better able to swing a sword [instead of
the strength gained from training barbell lifting improving your swing].

IMO, a skill-only system has a problem where there's not enough mutual
improvement. Animal charming is not distinctly related to haggling, but
both through practice will improve your general charisma to some degree,
which will have an effect on the other. Either the skill tree goes up to
one or several "root" skills, which are stats by another name, or it runs
the risk of eliminating the groupings of skills.

>
>Stats are just an abstraction, much more so than skills. You get
>stronger by lifting weights or rowing or other exercise, but which
>muscles are strong in what proportions depends on how you have
>exercised. If you do lots of word games and crossword puzzles, you'll
>do wonders for your verbal intelligence, but if you never juggle any
>numbers, your mathematical intelligence will rot. You can learn to
>juggle or you can learn to do slight-of-hand, but an aptitude for one
>use of 'dexterity' does not imply an aptitude for the other. I think
>feedback to stats from skills is a good step in the direction of
>realism, if that's what you want. Elimination might even be better,
>stats were a crude generalization of abilities created for class/level
>based gaming. A good skill set can define a character better. Some
>'aptitudes' makes sense. A character with the 'Bookish' aptitude would
>do better at academic skills, without implying that this means greater
>intelligence in general.

So you're advocating the replacement of half a dozen stats with about 30
narrowly-defined ones? A viable option, I suppose.

>
>>Add parry and dodge.
>
>Oh, yeah. Defensive skills are a must. Once you have a dodge skill,
>that's about 50% of the old DEX stat taken care of, right there.
>
>>>MAGIC:
>>>Spell power bonus (fire, cold, energy, teleportation, enchant, death)
>>>Concentration - Meditation - MP bonus
>>>Lower casting cost (MP)
>
>>For concentration, see the above.
>
>??? You want magic points to be purely stat based? If so, I disagree.
>Skill is important. Level is at least as important as spell stat in
>your standard roguelike. To truly do a skill-based game means ditching
>levels, which means skill has to do the work that level did. You have
>to buy those magic point advances now by developing the appropriate
>skill(s).

I wasn't exactly clear on this. I was interpreting concentration strictly
as he wrote it, a skill whose sole function was to provide an MP bonus.
The MP derived stat can be done entirely through Int/Wis/etc, and casting
costs can be discounted through proficencies in the individual spell, spell
college, and general magic (a magic skill tree, in essence).

>
>>I'd recommend grouping spells under general genres, like evocation,
>>divination, foo elemental, etc., and inheriting experience from individual
>>spells to the genre which provide a bonus to all spells in the genre.
>
>Depends on how many spells you have. Not worth it if you only have 50
>spells. And your approach creates complication in multiple type
>spells. Is summon fire elemental to use 'fire magic' skill or
>'summoning' skill? Or the game designer may want to have character
>race determine base skill in a magical type, and only allow skill to
>enhance particular skills.

Could you clarify that last sentence? I think you used about five too many
definitions of the word 'skill' there. :)

You may be right about summoning... summoning of mundane monsters could be
stuffed under the same spell genre as teleportation, as you're calling them
from somewhere and not making them from nothing.

Even with a mere 50 spells (and I'd love to see about a thousand or so.
:)), you could gain by creating fire/ice/air/earth elemental categories,
evocation, and divination. Evoocation handles the magic-missile, teleport,
and other pure-magic spells, and dvination gets identify, know alignment,
etc. You thus get related spells together under a single category, and can
assign a bonus/skill to the lot.


>>Lower casting costs would go with individual spell proficency.
>
>I agree on this point. Improved general efficiency is better modeled
>as increased spell points.

I only agree with you on global magic == improvement because you run the
risk of cutting off significant discounts due to rounding -- Fire bolt
costing 9.9 vs 9.1 is a pretty large difference, but they would both get
truncated to 9.

>>>MOVEMENT (World):
>>>Use transport - (Horse riding, boat)
>>>Portal control - allow use world portals
>>>Access area - Climbing, Swimming, Pathfinding
>>>Cartography - Map making - allow map?
>>>Bridge building
>
>A skill to double movement speed on a large-scale map only? Distance
>Running? Or that could be your Pathfinding?

Sure, trail-finding sounds like a reasonable skill.

>
>>Mount, bridge building, and the Access Area set would probably all go under
>>woodcraft.
>
>Woodcraft? Climbing & swimming are skills for the urban thief. Riding
>for the knight who never found his own food in his life. Bridge
>building is best done by a team of engineers with sound educations. A
>woodsman may be able to drop a log across a stream, but that's about
>it. For rivers, even small ones, you need to know what you are doing,
>and that isn't part of woodcraft. And horses are creatures of the open
>plain. The more wooded the area, the less useful they become.

Woodcraft was badly named here... what would you call a Ranger's general
set of skills and proficencies? All of the ones I listed had to do with
outdoor, wilderness activities.

>
>>"Allow use [of] world portals?" They're static objects, not living things
>>-- if you can throw a bowling ball in and have it come out on the other
>>side, a person should be able to use 'em without a skill check.
>
>On the other hand, maybe you could get lost if you don't know what you
>are doing. Or die. He obviously has something magical in mind, in
>which case it can work however he pleases.

True, although by most definitions of portals they're pretty simple things.
The bowling ball test works fine as an indicator, though.

>
>>>Teleport self
>>>Phasing walls (ignore walls)
>>>Swimming
>>>Direction sense (?)
>
>Like mapping, this is pretty much got to be an automatic. We can see
>which way is toward the top of the monitor and nothing else should
>matter. This would be an observation skill anyway, not movement.

The only way to really do a D&D-esque Direction Sense would be to randomly
rotate the map, which is just annoying. :)


>>Phase walls and teleport self are both spells.
>
>For *humans* phasing through walls and teleporting are spell effects.
>For quantum pixies, they may be skills.

I'd argue that the magical quantum pixies are still directing magic energy
towards the goal of phase wall & teleport self, so they're still casting
the spell in some sense, albeit with an insane proficency/effectiveness.


>>If, by "more furs" you mean being able
>>to skin more and better furs, that's skinning and is a woodcraft skill.
>
>When you mean 'general wilderness-type skill', please don't call it
>'woodcraft'. Much wilderness is not wooded.

Yeah, but I was unable to think of a short general-wilderness-type-skill
adjective at the moment.

>
>>Pick pocket is a thiefy skill.
>
>One of the reasons for going to skills system is to leave behind this
>kind of automatic stereotyping and open up character development to
>something more flexible than fighter/mage/priest-healer/thief
>stereotypes.

Yeah, but pick pocket is more closely related (mentally and physicially)
with pick lock than ride horse. I wasn't saying that it was only useful
for thief-type characters, I was saying it belonged in a group with the
other skills I called thiefy.

>
>>Identify item is either a spell or an
>>appraisal (as an appraisal is just a limited identification), your pick.
>
>Identifying and appraising are two different things. A bard would
>sooner recognize the fabled Mandolin of the Highly Annoying Minstrel.
>But he'd be less equipped than his burglar pal to appraise it, i.e. to
>judge its market value.

But a truly accurate appraisal requires an identification. A violin expert
is going to be able to identify a Stradavarius right out, but he won't be
able to give an accurate appraisal if he doesn't. Likewise, if I guve you
a crystal-shaped clear rock, you won't be able to tell me how much it's
worth unless you know whether it's glass, quartz, or diamond.


>>Healing is a valid skill, and I would say it goes under woodcraft because
>>you use local herbs & the like as tools.
>
>You are more likely to be using wood, cloth, and mud. Herbs are good
>for a variety of conditions, but spurting blood isn't one of them. A
>prudent fighter will learn the art of first aid whether he dwell in
>the city or woodland village.

I was thinking about healing in the additional sense of natural
painkillers, herbal remedies, and the like... first aid may deserve its own
seperate skill.

>
>>Health is a statistic called Constitution. See above on athletics and
>>concentration.
>
>Again, there really is no need to keep stats if you have a good skill
>system. You are much too willing to accept 'the way it has always been
>done' as the only way.

True, but I happen to like the demarcation between a character's innate
physical and mental abilities and a character's learned proficencies. No
amount of knowledge in your noggin is going to make that snakebite less
deadly if you don't do anything about it, and we already have a
healing/first-aid skill for treatment.

I also don't think that the thread-beginning post said that it was a
skill-only RPG, so duplication of description in skills & stats something
to avoid.

>
>>Enchanting and alchemy both seem to belong primarially under magic.
>
>But they don't have to be treated as *spells*, so they are better
>grouped with other crafts than with spell-casting unless they *are*
>treated as spells. In many fantasy worlds and myths, creation of an
>enchanted sword is the ultimate power of smithing. At a certain high
>point of skill, a crafter becomes able to put a bit of himself into
>his work, creating a 'magic' item.

Well, I'm not making an exact differentiation between spells and skills --
a skill tree, for example, may work like:

Magic -> Evocation -> Magic Missile

or

Magic -> Enchantment -> Enchant Necklace/Create Wand/Create Ring/Etc.

They both use the same general knowledge of magic, as in how to work and
use magical energies, but they differ in that the former is interested in
creating an immediate, pure magical effect (evocation -> magic missile)
whereas the latter is interested in creating a largely permanant (or at
least stored) effect on an item.

>
>>Pet control and leadership are both things you get from Charisma, so no
>>skill for them.
>
>Right. Nobody ever learned anything about handling animals or dealing
>with people. You're born with it, or don't even bother. Sure.

I'll admit this was one of my weaker points, although isn't leadership a
definition for Charisma?

>
>>Learning is intelligence, not a skill.
>
>I disagree with your phrasing (because you appear to be basing this on
>stat-centered thinking), but I agree that this shouldn't be a skill. A
>general experience boost from a skill is likely to be seriously
>abusable. However, I could see a Studying skill that would improve
>your ability to absorb training. Good study habits do matter. And a
>high Studying skill could make it easier to find a teacher, as that
>wizard may not want to pass on his arcane knowledge of the Making
>Pointy Sticks Fly At Your Enemies spell if he thinks he might be
>wasting his time.

True, but Study is the knowledge of how to maximize learning effort,
whereas Intelligence is the actual ability to learn and make logical
connections. Study would be a valid skill if you could differentiate it
enough from Intelligence and Concentration.

Michael Blackney

unread,
Aug 20, 2001, 3:26:22 AM8/20/01
to
R Dan Henry <rdan...@earthlink.net> wrote...

>
> On Thu, 16 Aug 2001 04:04:15 GMT, the disembodied brain of
> pfh...@gdi.net (Chris Subich) transmitted thus:
>
> >On Wed, 15 Aug 2001 22:25:08 +0200, Archibald <kko...@kki.net.pl>, the
> >evil little half-monkey he is, wrote:
> >
> >>What skills we can use in RL game? If you know more, add it to the list.

I don't like to admit it but I like this idea. If you were given some
numbers based on total skills in certain areas I'd feel much more
comfortable.


> >>MOVEMENT (Dungeon):
> >>Searching
> >>Detect traps
>
> These really aren't movement skills. These are observation spells.

Observation skills. If you can do it in the real world it does not
require magic. Of course if you're looking for a needle in a
haystack...

>
> >>Stealth
> >>Pick lock (doors)
>
> If you can pick a lock on a door, I'll bet you can pick a lock on a
> chest, too. :-) Wouldn't call this movement, myself.

Not necesarily. Locks on doors are regularly trivial (unless the door
is to a treasure room). Semantics aside, I think that chests should
be *much* harder to open than doors.

> >Searching [detect traps and secret doors], hide, move silently, and pick
> >lock are all thiefy skills.
>
> I disagree with the stereotyping. Hiding, for example, is a good skill
> for anyone. And if you have a woodcraft skill area, moving silently in
> the wilderness would go there.

I thought it should go in the wilderness skill area, after all not all
wilderness is wooded. <g>

>
> >>TREASURE:
> >>more artifacts/rare items
>
> Crafting/repair abilities cover this.

An item/artefact lore skill might not go astray. The idea that
characters know a little about the history of their world works well
with me. This information should all be available to the player, &
could influence the player's actions in several ways.

> >Pick pocket is a thiefy skill.
>
> One of the reasons for going to skills system is to leave behind this
> kind of automatic stereotyping and open up character development to
> something more flexible than fighter/mage/priest-healer/thief
> stereotypes.

Though I don't think it's healthy to ignore all forms of
categorisation simply to avoid stereotyping - some skills are related,
this we cannot deny. It may be best if skills can associate more
freely with each other (who was it who suggested a skill 'web' last
year?) than basic categories.

Michael Blackney

Archibald

unread,
Aug 20, 2001, 6:49:40 PM8/20/01
to
> >What skills we can use in RL game? If you know more, add it to the list.
>
> What kind of Roguelike are you going for? Will you use a skill-tree
> method?

Well, this is only *list of skills*. Nothing more :). I want to know
which skills we can use in RL games. Which to choose is completly
different question. Of course there can be doubled skills or half-doubled
skills or skills which can be even implemented as an ability (IMO ability
is a skill which is used very often and/or which is used as bonus skill
for other skills). In this moment this is not important. This is only
list of skills :).


Archibald

R Dan Henry

unread,
Aug 25, 2001, 4:07:32 AM8/25/01
to
On 20 Aug 2001 00:26:22 -0700, the disembodied brain of
michael...@hotmail.com (Michael Blackney) transmitted thus:

>R Dan Henry <rdan...@earthlink.net> wrote...
>>
>> On Thu, 16 Aug 2001 04:04:15 GMT, the disembodied brain of
>> pfh...@gdi.net (Chris Subich) transmitted thus:
>>
>> >On Wed, 15 Aug 2001 22:25:08 +0200, Archibald <kko...@kki.net.pl>, the
>> >evil little half-monkey he is, wrote:
>> >
>> >>What skills we can use in RL game? If you know more, add it to the list.
>> >Statistics, as fundamental character attributes,
>> >should provide boni to skills, and not the other way around.
>>
>> Why? Sure, this is the way it's normally been done, but it doesn't

[snip]

>> Stats are just an abstraction, much more so than skills. You get
>> stronger by lifting weights or rowing or other exercise, but which
>> muscles are strong in what proportions depends on how you have
>> exercised. If you do lots of word games and crossword puzzles, you'll
>> do wonders for your verbal intelligence, but if you never juggle any
>> numbers, your mathematical intelligence will rot. You can learn to
>> juggle or you can learn to do slight-of-hand, but an aptitude for one
>> use of 'dexterity' does not imply an aptitude for the other. I think
>> feedback to stats from skills is a good step in the direction of
>> realism, if that's what you want. Elimination might even be better,
>> stats were a crude generalization of abilities created for class/level
>> based gaming. A good skill set can define a character better. Some
>> 'aptitudes' makes sense. A character with the 'Bookish' aptitude would
>> do better at academic skills, without implying that this means greater
>> intelligence in general.
>
>I don't like to admit it but I like this idea. If you were given some
>numbers based on total skills in certain areas I'd feel much more
>comfortable.

No way to give numbers without working out the system. I mean, without
a game mechanic decided upon, a skill of 50 might be average,
terrible, or godlike. But look at any existing skills-centered system
and just eliminate stat mods. Now look at a possibility of aptitudes,
which some skill-and-stat systems already have. Each 'aptitude' would
give a bonus to a defined list of skills appropriate to the theme.
'computer geek' and 'researcher' aptitudes would both get a bonus to
'search internet' skill, for example. Bonus could be in the form of a
set figure or, if levels are retained, in the form of a per level
bonus. Personally, I prefer a modifier that speeds advancement in that
skill. For example, if skill increase becomes more difficult at higher
skills, someone with an aptitude that benefited that skill would be
treated as having a lower skill for advancement purposes.

A simple example of the last bonus method above. My Mutant Mantis
Accountant with the Computer Geek aptitude is searching the net for
encryption programs to protect his embezzlement records. He succeeds
and a skill-advancement check is made. The game uses a simple 0-100
rating for skills where skill = % chance of success. (This is very
crude, but this is just an example.) My Mantis has a Search Internet
skill of 98% and the check for skill advancement is d100 > current
skill, so normally I'd have a feeble 2% chance of moving up to skill
99%. However, thanks to an aptitude for this skill, my skill is
treated as current skill minus 10 or 88%. My chance is a much better
12%. Of course, at skill = 100%, no check is made and under this
simple system a skill of 101% wouldn't be more useful anyway.

>> >>MOVEMENT (Dungeon):
>> >>Searching
>> >>Detect traps
>>
>> These really aren't movement skills. These are observation spells.
>
>Observation skills.

Yes. Simple error on my part. I meant skills.

>> >>Stealth
>> >>Pick lock (doors)
>>
>> If you can pick a lock on a door, I'll bet you can pick a lock on a
>> chest, too. :-) Wouldn't call this movement, myself.
>
>Not necesarily. Locks on doors are regularly trivial (unless the door
>is to a treasure room). Semantics aside, I think that chests should
>be *much* harder to open than doors.

Yes, in general. But it is still the same *skill*, just a harder skill
test.

>> >Pick pocket is a thiefy skill.
>>
>> One of the reasons for going to skills system is to leave behind this
>> kind of automatic stereotyping and open up character development to
>> something more flexible than fighter/mage/priest-healer/thief
>> stereotypes.
>
>Though I don't think it's healthy to ignore all forms of
>categorisation simply to avoid stereotyping - some skills are related,
>this we cannot deny. It may be best if skills can associate more
>freely with each other (who was it who suggested a skill 'web' last
>year?) than basic categories.

I'm not against categories, if they're based on some intrinsic
similarity. I am against grouping skills simply because they are
particularly useful together when playing certain character 'types'.

The average (stage) magician is probably a much better pick pocket
than the average thief, as most thieves are not pickpockets (it's a
specialty, after all, and a somewhat tricky one) and do not practice
skills as closely related to the skill of picking pockets as is
slight-of-hand.

--
R. Dan Henry, Emperor of the Universe
rdan...@earthlink.net

Please Send Me All Your Unwanted Money

R Dan Henry

unread,
Aug 25, 2001, 6:41:44 AM8/25/01
to
Warning, this is quite long. I have had to keep most of the previous
material to maintain context for this rather elaborate discussion. I
hope it all makes sense, as I'm writing in the middle of the night.

On Mon, 20 Aug 2001 03:22:17 GMT, the disembodied brain of


pfh...@gdi.net (Chris Subich) transmitted thus:

>On Mon, 20 Aug 2001 00:30:24 GMT, R Dan Henry <rdan...@earthlink.net>,


>the evil little half-monkey he is, wrote:
>
>>On the contrary, having a skill to improve critical hit chances allows
>>you to have high-skilled but wild-swinging Barbarian-Dervishes and
>>high-skilled, precise Expert Duelists in your world. The former would
>>develop the ability to hit consistently and hit hard, whereas the
>>later would develop the ability to hit consistently and to find the
>>weaknesses in an opponents defenses. The difference between 2d6+4
>>every turn with only lucky crits and 2d6 plain, but frequent crits.
>>This makes more sense if crits don't just add damage, but have other
>>effects, like bypassing armor or bypassing random hit location to hit
>>a vital area. Precision is a good name for this skill and if it gets
>>big bonuses on unaware targets, it is Assassination, too.
>
>An Expert Duelist, however, is not going to be toting a two-handed
>broadsword, and a barbarian is not going to bt playing with the sharp
>pointy stick that other people call rapiers.

Well, they won't do well with them, but each might use a spear. The
fact that certain weapons don't lend themselves well to certain styles
of fighting doesn't negate the fact that most weapons can be used in a
variety of ways. If you make the critical-controlling stat on a weapon
a skill modifier (eg, a weapon uses anywhere from 20% to 150% of a
character's skill in the critical hit calculation), then you still
show the difference between the two fighting styles with all weapons
and the inherent 'preferred style' of the weapons. The
Barbarian-Dervish will score criticals slightly more often with a
rapier (if his skill is not so low the rapier bonus is rounded down to
nothing) than with his normal greatsword, but won't be comfortable
enough with it to get the sort of results the Expert Duelist gets.
Whereas the Expert Duelist with gets more criticals than the B-D, even
using a greatsword, but he'll do far less well than with a weapon that
lets him use his skill more effectively.

>D&D has critical range based upon weapon type, and I happen to aggree with
>that -- perhaps allow existing combat boni to come into play [say the
>weapon has a critical hit ratio, and you criticially hit when your roll +
>boni is more than ratio times defender's roll + boni].

Strange that you say 'D&D' has this. This is a crude version of
systems many games had 20 years ago.

>In my mind, skills should be something definite. In the case of combat,
>it's something you do -- swing a broadsword or dodge. A critical hit, to
>me, is a lucky or expert result from a hit -- you usually aren't aiming to
>do a critical hit over and above aiming to hit and cause damage to your
>enemy. A critical hit is a subset of a hit, so I don't think it needs its
>seperate skill.

Well, swordfighting is a subset of fighting, so why have multiple
combat skills?

If you don't want to use a skill, don't. But don't try to deny its
legitimacy.

>>>An athletics skill, by providing statistic boni, conflicts with separation
>>>of skills and statistics.
>>
>>Better complain about body-building, too, then, it's taking over CON.
>
>That was a listed skill? I missed it.

Yep. For hit point development. Of course, this doesn't even require
getting rid of stats. Rolemaster required you to spend development
points to increase concussion hits, but had a CON stat. But Rolemaster
is a real rpg, and getting rid of stats in a real rpg is difficult
because they give you something to work with in situations outside the
rules. With a roguelike, you don't need anything like that, because
everything that can happen must be explicitly programmed and a skill
can be assigned.

>>
>>>Statistics, as fundamental character attributes,
>>>should provide boni to skills, and not the other way around.
>>
>>Why? Sure, this is the way it's normally been done, but it doesn't
>>have to be. You really don't even need to have stats in skill-based
>>game. A strong character is one with high brute force skills, a
>>dexterous character is one with lots of acrobatic skills, a
>>charismatic character is one with lots of social skills. It isn't as
>>if stats are pristine and sacred in roguelikes, anyway. Most, if not
>>all, developed roguelikes involve a process of maxing out your stats,
>>same as developing skills. In Nethack you can even do some of this by
>>practicing the stat, just like practicing a skill to improve. In real
>>life, practicing relevant skills boosts the stat.
>
>True on the last point, but that's a function of the skill making use of
>the stat. Simply knowing athletics does not make me stronger, it's the use
>of it (which also makes use of the stats involved) that makes me stronger.

But you are not 'strengthing', you are 'rowing' or 'lifting' or
'carrying'. You are doing some particular thing that builds strength.
And 'knowing athletics' won't improve your Basketball skill any more
than it will your Strength. That you must actually practice the skill
to get the benefit is not relevant.

>I really happen to like the skill tree idea -- stats up at the very top,
>and skills make use of stats to modify their rolls, but the simple presence
>or absence of a skill does not change the stat. This also prevents
>unwanted codependency between independent stats, e.g. where your high skill
>score im barbell lifting makes you better able to swing a sword [instead of
>the strength gained from training barbell lifting improving your swing].

No, you just want the strength gain from lifting barbells to
miraculously improve my damage bonus when I kick. The only difference
between the two cases you give above is that one adds a layer of
indirection which the other does not. If the relationships are hidden
from the player, he'll never see any difference.

>IMO, a skill-only system has a problem where there's not enough mutual
>improvement. Animal charming is not distinctly related to haggling, but
>both through practice will improve your general charisma to some degree,
>which will have an effect on the other.

If you want that, it's easy enough to use a skill tree. But I disagree
that these should cross-pollenate. Good with animals and good with
people are quite different.

>Either the skill tree goes up to
>one or several "root" skills, which are stats by another name, or it runs
>the risk of eliminating the groupings of skills.

I disagree that the top of a skill tree is a stat. Especially as games
can and do have both skill trees and stats that are not part of the
skill tree. Stats are generally treated quite differently from skills,
whereas the top of the skill tree is just another skill that happens
to not have another more general skill above it. And while stats
pretend to be some basic fact about a character, maybe even a genetic
one (although they're all too broad for that), the top of a skill tree
represents experience and education.

>>realism, if that's what you want. Elimination might even be better,
>>stats were a crude generalization of abilities created for class/level
>>based gaming. A good skill set can define a character better. Some
>>'aptitudes' makes sense. A character with the 'Bookish' aptitude would
>>do better at academic skills, without implying that this means greater
>>intelligence in general.
>
>So you're advocating the replacement of half a dozen stats with about 30
>narrowly-defined ones? A viable option, I suppose.

No, because 'aptitudes' would be binary qualities. And they could
overlap whenever appropriate. They function quite differently from
stats. I just gave some detail in another post, so I'll not go into it
here.

>>>>MAGIC:
>>>>Spell power bonus (fire, cold, energy, teleportation, enchant, death)
>>>>Concentration - Meditation - MP bonus
>>>>Lower casting cost (MP)
>>
>>>For concentration, see the above.
>>
>>??? You want magic points to be purely stat based? If so, I disagree.
>>Skill is important. Level is at least as important as spell stat in
>>your standard roguelike. To truly do a skill-based game means ditching
>>levels, which means skill has to do the work that level did. You have
>>to buy those magic point advances now by developing the appropriate
>>skill(s).
>
>I wasn't exactly clear on this. I was interpreting concentration strictly
>as he wrote it, a skill whose sole function was to provide an MP bonus.

So was I. I don't see why you think this can't be done. Again,
Rolemaster was a very skill-based system. It began granting automatic
spell point increases each level, but with Rolemaster Standard System,
it moved to requiring developing spell points. It works. You may
personally dislike it, but it is perfectly workable as a mechanic.

>The MP derived stat can be done entirely through Int/Wis/etc, and casting
>costs can be discounted through proficencies in the individual spell, spell
>college, and general magic (a magic skill tree, in essence).

It certainly *can* be done entirely through stats, but this thread
didn't start out with someone asking what can be done *without*
skills, the question you seem interested in addressing, but in what
can be done *with* them. The answer is, I think, almost anything you
can imagine a use for.

>>Depends on how many spells you have. Not worth it if you only have 50
>>spells. And your approach creates complication in multiple type
>>spells. Is summon fire elemental to use 'fire magic' skill or
>>'summoning' skill? Or the game designer may want to have character
>>race determine base skill in a magical type, and only allow skill to
>>enhance particular skills.
>
>Could you clarify that last sentence? I think you used about five too many
>definitions of the word 'skill' there. :)

You have a skill tree, a simple two level one with each 'type' of
magic on the top and a set of individual spell skills underneath it.
So, you have Enchantment skill and Runic Blade spell skill. Anyone can
learn the Runic Blade spell and develop skill with it, but the skill
check to see how much of a bonus is given by the Runic Blade
enchantment is made against Enchantment skill + Runic Blade skill in a
typical skill tree method. But that doesn't mean that Enchantment
skill itself has to be trainable (the magic model may involve each
spell having a 'uniqueness' that prevent learning from one spell
applying to another, which explains why the magic type skill is a
fixed talent rather than a standard skill). An elf may have
Enchantment 30 and a human only Enchantment 5 and the only way for a
human to get that 30 is to change into an elf.

I wouldn't do it this way, but it is an option.

>You may be right about summoning... summoning of mundane monsters could be
>stuffed under the same spell genre as teleportation, as you're calling them
>from somewhere and not making them from nothing.

It isn't a point about summoning. It's that spells don't naturally
fall into exclusive categories and don't work that well with rigid
skill trees. You need to make some arbitrary decisions as to which
tree to put it under, allow identical spells under multiple trees, or
allow trees to 'cross' and have spells that belong in multiple trees
and interact with multiple skills on the next level up. For my new,
non-summoning example, a Steam Bolt spell -- fire and water magic
both. *Really* makes your system tough if 'opposite' element magic
types have some form of 'conflict' going on.

>Even with a mere 50 spells (and I'd love to see about a thousand or so.

I'd like to see whichever is more appropriate to the world-vision of
the game creator. The right 50 spells will go a long way and a
thousand spells give you a thousand chances to have one that ruins
game balance, so I can't go with a straight 'more is better'.

>:)), you could gain by creating fire/ice/air/earth elemental categories,
>evocation, and divination. Evoocation handles the magic-missile, teleport,
>and other pure-magic spells,

I gag at the idea of putting magic missile and teleport together.
Offensive spells belonging with teleport are things like 'displace
opponent' (teleport away, preferable somewhere nasty, 'flesh scramble'
(teleports parts of the body into different positions, thereby making
said body uncomfortable, and 'structural displacement' (teleport away
supporting bits holding up ceiling over foe's head). Magic missile is
directing raw magical energy and has more in common with enchanting or
disenchanting an item (or better, with charging a wand), than with
teleportation.

>etc. You thus get related spells together under a single category, and can
>assign a bonus/skill to the lot.

Yes, but as I said, there are spells that cross categories and you
need to deal with that if you use this method.

Anyway, we both agree that such skills are possible, which settles the
matter as far as the question of this thread is concerned. You just
seem to think they are more necessary and less problematic than I do.

>>>Lower casting costs would go with individual spell proficency.
>>
>>I agree on this point. Improved general efficiency is better modeled
>>as increased spell points.
>
>I only agree with you on global magic == improvement because you run the
>risk of cutting off significant discounts due to rounding -- Fire bolt
>costing 9.9 vs 9.1 is a pretty large difference, but they would both get
>truncated to 9.

So? You know you get rounding 'error' at some point no matter what
your system and you tweak the values accordingly. And I was only
talking global magic anyway "Improved ***general*** efficiency" --
proficiency with particular spells can be modeled by lowered casting
costs and/or improved effectiveness.

>>A skill to double movement speed on a large-scale map only? Distance
>>Running? Or that could be your Pathfinding?
>
>Sure, trail-finding sounds like a reasonable skill.

I dislike that name, simply because the skill should be useable even
in areas where there are not likely to *be* trails, eg deserts, or
finding trails requires no skill, eg you are using a main road.
Trained individuals can move faster, it isn't just a case of finding a
better path. I like Distance Running or Traveling.

>Woodcraft was badly named here... what would you call a Ranger's general
>set of skills and proficencies? All of the ones I listed had to do with
>outdoor, wilderness activities.

I'd call it a set of skills that happen to be useful in the same place
and are traditionally given to the stock game-fantasy character type
called 'ranger'. And then I'd put the observation skills into my
Observation skill tree and so on. I classify skills by what they are,
not by which stereotype is most associated with them.

If you want to model familiarity with a particular environment, like
the woods or cities or swamps, make that a skill. Your Rat-Man Skulker
and your Wood Elf Ranger both have a nice high Stealth skill of 70,
but the Skulker has an Urban Environment skill of 30 and the Ranger
likewise has a Woodlands Environment skill of 30, so while each is
pretty damn stealthy wherever he goes, the Skulker will be even better
in the city 70 base skill + 30 environmental familiarity = 100
effective skill and the Elf will be likewise incredible in the woods.
You'll also need to assign a value to each of the non-environmental
skills to indicate to what degree they depend on environment. Picking
Nose would get a 0 value, indicating that you can easily find your
nose wherever you are. Likewise, spell skills mostly would not involve
environment. Fighting skills would use part of the environmental bonus
(half, let us say), as you can use your environment to aid you in
combat, but it isn't normally a huge factor. For things like Stealth
or many Lore skills or Observation skills (if you don't know what
things *should* look like, it's harder to spot what's out of place),
the full value would be used. You could even give negative base skills
to certain races. Hawkmen might be naturally claustrophobic and start
with -20 to Underground Environment skill and thus do poorly in
dungeon settings, at least until they can acclimate themselves to an
alien setting.

>True, although by most definitions of portals they're pretty simple things.
>The bowling ball test works fine as an indicator, though.

Not always. Sometimes the challenge is one a bowling ball wouldn't
face, like a risk of insanity. It's *very* world-dependant to have a
skill specific to using magical portals, tho.

>>Like mapping, this is pretty much got to be an automatic. We can see
>>which way is toward the top of the monitor and nothing else should
>>matter. This would be an observation skill anyway, not movement.
>
>The only way to really do a D&D-esque Direction Sense would be to randomly
>rotate the map, which is just annoying. :)

Cool, something else we can agree on. Direction Sense in a roguelike =
bad.

>>>Phase walls and teleport self are both spells.
>>
>>For *humans* phasing through walls and teleporting are spell effects.
>>For quantum pixies, they may be skills.
>
>I'd argue that the magical quantum pixies are still directing magic energy
>towards the goal of phase wall & teleport self, so they're still casting
>the spell in some sense, albeit with an insane proficency/effectiveness.

Not necessarily with an insane proficiency. Starting skill could be
quite low. The difference between spell and skill is that they need
not meet requirements for spell-casting (some games don't allow spell
casting if blinded, silenced, confused, etc. or require some item,
typically a book, in order to cast a spell) in order to use the skill
and do not need spell points to use this natural ability.

A Sea-Troll might reason as you do about us and decide we cast
"Breathe Air" with an insane proficiency and must all be spell-casters
for all that we deny it.

Magic != Spell

>>When you mean 'general wilderness-type skill', please don't call it
>>'woodcraft'. Much wilderness is not wooded.
>
>Yeah, but I was unable to think of a short general-wilderness-type-skill
>adjective at the moment.

Wilderness Skill? Outdoor Skill? Nature skill? How about 'Ranging'
since you associate these skills with Rangers? :-)

>>>Pick pocket is a thiefy skill.

>>One of the reasons for going to skills system is to leave behind this
>>kind of automatic stereotyping and open up character development to
>>something more flexible than fighter/mage/priest-healer/thief
>>stereotypes.

>Yeah, but pick pocket is more closely related (mentally and physicially)
>with pick lock than ride horse.

That why it would be a Manipulation skill and be on a different skill
tree than riding. I'm not sure I wouldn't make a Riding skill at the
top of its own skill tree and do animal types as particular skills.
Riding is actually one of the more difficult skills to assign to a
larger category as it can reasonably be assigned to several -- Animal
skills, Athletic skills, Acrobatic skills, Butt Impact Endurance
skills :-)... Outdoor/Wilderness skills if you insist on having such a
category.

>I wasn't saying that it was only useful
>for thief-type characters, I was saying it belonged in a group with the
>other skills I called thiefy.

I disagree that these make a natural grouping. Sneaking around is a
'thiefly' art, but it has little to do with picking pockets. Nor does
either have much to do with picking locks. You associate these skills
because they are often useful to the same people, but that is a poor
reason to put them in the same skill tree. If you want to mimic the
training of traditional fantasy stereotypes, include some sort of
Guild Training that let's you start with a set of relatively high
starting skills appropriate to some stereotype.

>>>Identify item is either a spell or an
>>>appraisal (as an appraisal is just a limited identification), your pick.
>>
>>Identifying and appraising are two different things. A bard would
>>sooner recognize the fabled Mandolin of the Highly Annoying Minstrel.
>>But he'd be less equipped than his burglar pal to appraise it, i.e. to
>>judge its market value.
>
>But a truly accurate appraisal requires an identification. A violin expert
>is going to be able to identify a Stradavarius right out, but he won't be
>able to give an accurate appraisal if he doesn't. Likewise, if I guve you
>a crystal-shaped clear rock, you won't be able to tell me how much it's
>worth unless you know whether it's glass, quartz, or diamond.

Right, before you can do an appraisal, you need to identify the
object. It's still too different processes and two different skills.
And you don't need to know if it is *really* the fabled Ring of the
Jeweler King in order to appraise it... as long as your mark can be
convinced that it is.

>I was thinking about healing in the additional sense of natural
>painkillers, herbal remedies, and the like... first aid may deserve its own
>seperate skill.

There's still not much you can do for injuries with herbs that you
can't do without. Painkillers are pretty much it. And that's a matter
of chewing on (or whatever) a prepared dose you got before hand, as
trying to find a particular plant while in severe pain and suffering
blood-loss is not likely to succeed no matter how skilled you are.

>>>Health is a statistic called Constitution. See above on athletics and
>>>concentration.
>>
>>Again, there really is no need to keep stats if you have a good skill
>>system. You are much too willing to accept 'the way it has always been
>>done' as the only way.
>
>True, but I happen to like the demarcation between a character's innate
>physical and mental abilities and a character's learned proficencies.

There are no *innate* abilities. Potentialities, maybe, but not
abilities. Anyway, you're free to use stats if you wish, but the
question asked was what *can* be done *with* skills. You keep talking
about what can be done *without* skills.

>No
>amount of knowledge in your noggin is going to make that snakebite less
>deadly if you don't do anything about it, and we already have a
>healing/first-aid skill for treatment.

That's the difference between having a skill and using it, nothing to
do with skills vs. stats. The only thing that protects from snakebite
after the fact is body mass. Specific health problems can make it
deadlier, but there really is no "Constitution" that will make the
poison go away. And Strength, as a measure of muscle mass is a better
indicator of how much of a beating you take. Certainly, at some points
using skills involve some heavy abstraction from reality, but stats
involve even more, IMO.

>I also don't think that the thread-beginning post said that it was a
>skill-only RPG, so duplication of description in skills & stats something
>to avoid.

No, but it didn't say that. But it didn't say that it *wasn't*,
either. Re-offering a much-used model adds less, I think, to the
conversation than exploring the limits of the question actually asked.
What skills *can* we have? Well, you can have a skill for anything in
the game.

>>>Enchanting and alchemy both seem to belong primarially under magic.
>>
>>But they don't have to be treated as *spells*, so they are better
>>grouped with other crafts than with spell-casting unless they *are*
>>treated as spells. In many fantasy worlds and myths, creation of an
>>enchanted sword is the ultimate power of smithing. At a certain high
>>point of skill, a crafter becomes able to put a bit of himself into
>>his work, creating a 'magic' item.
>
>Well, I'm not making an exact differentiation between spells and skills --
>a skill tree, for example, may work like:
>
>Magic -> Evocation -> Magic Missile
>
>or
>
>Magic -> Enchantment -> Enchant Necklace/Create Wand/Create Ring/Etc.
>
>They both use the same general knowledge of magic, as in how to work and
>use magical energies, but they differ in that the former is interested in
>creating an immediate, pure magical effect (evocation -> magic missile)
>whereas the latter is interested in creating a largely permanant (or at
>least stored) effect on an item.

And I'm saying it doesn't have to be in the magic skill tree at all!
It is perfectly in keeping with various fantasy works and their
original mythic inspirations to having it on this line:

Metalworking -> Blacksmithing -> Weaponsmithing

and have a high enough Weaponsmithing skill start turning out magical
weapons. Armorsmithing right next to it on the skill tree makes the
magical armors. Alchemy under whatever you want to call general
working-with-ingredients skill tree which would also include Cooking,
Working With Poisons, Preparing Herbal Remedies, etc. would probably
branch into Analysis, Transformations, Preservation. Something like
that. Transformations would involve turning ingredients into a new
thing and would make magical potions at higher levels. Magical
writings (runes) would involve very high calligraphy or maybe
wood-carving if the runes were carved (remember that from the
primitive point of view where these myths come from, the *words
themselves* carry magical power so language manipulation that seems
ordinary to us was enchantment to them). Some items may be the work of
spells, but the can easily mostly be the work of the 'supernaturally'
high level of skill of master craftsmen, whose ability is so great
that it transfers their own spiritual power into the work, making
magic.

Obviously, the desirability of this is highly world dependant. It'd be
great for Middle-Earth, where most elvish and all dwarven magic seems
to work this way. It'd be terrible for Conan's Hyborian Age or Elric's
Young Kingdoms setting. But it's a perfectly workable system.

>>>Pet control and leadership are both things you get from Charisma, so no
>>>skill for them.
>>
>>Right. Nobody ever learned anything about handling animals or dealing
>>with people. You're born with it, or don't even bother. Sure.
>
>I'll admit this was one of my weaker points, although isn't leadership a
>definition for Charisma?

Leadership isn't an inherent trait, although certain inherent traits
help greatly -- height, for example, is a clear winner. But leadership
isn't even a single skill. I wouldn't try to make it a skill. My idea
of a skill tree for Social Skills would look something like this if
you wanted three levels.

Social Skills----Humans-------Bribery
\ \\_Haggling
| \_Befriending, etc.
|_Elves--------Bribery, etc.
\Trolls-------Bribery, etc.

Leadership would involve, first, Befriending someone, so that you were
seen positively. Then a Bribery attempt could be made to hire that
person on, or a Scamming attempt made to convince the poor dupe to
join up for free ("You'll see the world!"). Later, some skill I can't
think of a name for at the moment would be used to inspire your
followers to boost their morale (Ah, call it Inspiration!) In short,
leadership will depend on a slew of social skills as well as good
judgment -- if you insist on leading your Lotus-Pixie Gossips against
a legion of Dragon-Troll Reavers, well, your Inspiration skills won't
matter much.

>>>Learning is intelligence, not a skill.
>>
>>I disagree with your phrasing (because you appear to be basing this on
>>stat-centered thinking), but I agree that this shouldn't be a skill. A
>>general experience boost from a skill is likely to be seriously
>>abusable. However, I could see a Studying skill that would improve
>>your ability to absorb training. Good study habits do matter.
>

>True, but Study is the knowledge of how to maximize learning effort,
>whereas Intelligence is the actual ability to learn and make logical
>connections.

Which is a learned skill. Actually, it's several learned skills. There
are some apparently innate tendencies and potentials, but the closest
thing I can think of to a general intelligence trait is curiosity --
and even that usually has developed some focus by adulthood. The
ability that derives from this sense of interest is exactly what is
modeled by my 'aptitudes'. Now, a 'Bookish' aptitude will probably be
judged a sign of more intelligence than a 'Car Guy' aptitude, but it's
the same thing applied to a different field.

>Study would be a valid skill if you could differentiate it
>enough from Intelligence and Concentration.

Sure, Intelligence is a huge abstraction that isn't in the game at all
under the model I've been offering, so that's easy. Concentration is
just a spell point enhancer according to the skill list given, so that
one is easy. If you want a more natural definition for Concentration
skill I'd call it 'the ability to maintain focus on one thing for a
sustained period' and use it for maintaining spells where wavering
one's focus will drop the spell or for staring contests or stuff like
that... immediate benefits. Concentrating on a text doesn't do much
good if you don't know how to effectively process what you read,
whereas if you know what you are doing you *can* learn by spending
half an hour in the library, going out and tossing the frisbee for
half an hour, and then studying some more. If you can't concentrate at
all, you won't be able to study, no. That, however, is a question of
pathology, not skill. You don't have to be a master concentrator to be
a master studier.

--
R. Dan Henry, Emperor of the Universe
rdan...@earthlink.net

Chris Subich

unread,
Aug 25, 2001, 11:42:19 AM8/25/01
to
On Sat, 25 Aug 2001 10:41:44 GMT, R Dan Henry <rdan...@earthlink.net>,
the evil little half-monkey he is, wrote:

>>An Expert Duelist, however, is not going to be toting a two-handed
>>broadsword, and a barbarian is not going to bt playing with the sharp
>>pointy stick that other people call rapiers.
>
>Well, they won't do well with them, but each might use a spear. The
>fact that certain weapons don't lend themselves well to certain styles
>of fighting doesn't negate the fact that most weapons can be used in a
>variety of ways. If you make the critical-controlling stat on a weapon
>a skill modifier (eg, a weapon uses anywhere from 20% to 150% of a
>character's skill in the critical hit calculation), then you still
>show the difference between the two fighting styles with all weapons
>and the inherent 'preferred style' of the weapons. The

That might work, but I still don't like the idea of a critical hits skill.

In my mind, an end-level skill is something that either can be actively
used or isn't terribly hard to practice. For example, Fire Magic makes an
acceptable skill because you can go off and cast Fireball to your heart's
content. Likewise, Sword Style (of fighting) makes sense as a skill
because all you have to do to get better at it is go off and hack apart a
few bears. How do you propose that a PC set about practicing his critical
hits skill?

If, as you say, critical hits are a function of fighting style, then the
duelist is improving his proficency in that particular fighting style [of
rapid, accurate hits], not just a random critical hit skill. Likewise, the
barbarian improves his berzerker fighting style.

>
>>D&D has critical range based upon weapon type, and I happen to aggree with
>>that -- perhaps allow existing combat boni to come into play [say the
>>weapon has a critical hit ratio, and you criticially hit when your roll +
>>boni is more than ratio times defender's roll + boni].
>
>Strange that you say 'D&D' has this. This is a crude version of
>systems many games had 20 years ago.

... and I would know this because of my excellent gameplaying skills when I
wasn't yet in elementary school.

>
>>In my mind, skills should be something definite. In the case of combat,
>>it's something you do -- swing a broadsword or dodge. A critical hit, to
>>me, is a lucky or expert result from a hit -- you usually aren't aiming to
>>do a critical hit over and above aiming to hit and cause damage to your
>>enemy. A critical hit is a subset of a hit, so I don't think it needs its
>>seperate skill.
>
>Well, swordfighting is a subset of fighting, so why have multiple
>combat skills?

Because there is (usually) a logical division between swordfighting and
spear-fighting, and I can deliberately set out to practice either.


>>>>An athletics skill, by providing statistic boni, conflicts with separation
>>>>of skills and statistics.
>>>
>>>Better complain about body-building, too, then, it's taking over CON.
>>
>>That was a listed skill? I missed it.
>
>Yep. For hit point development. Of course, this doesn't even require
>getting rid of stats. Rolemaster required you to spend development
>points to increase concussion hits, but had a CON stat. But Rolemaster
>is a real rpg, and getting rid of stats in a real rpg is difficult
>because they give you something to work with in situations outside the
>rules. With a roguelike, you don't need anything like that, because
>everything that can happen must be explicitly programmed and a skill
>can be assigned.

Yes, but stats make an excellent foundation for skills. If I drink a
potion of strength and get stronger, I'll hit harder, row faster, and break
down more doors. The potion made me stronger -- it didn't provide a bonus
to swordfighting, rowing, and door bashing.

In implementation terms, a skill tree with stats used as modifiers makes
for a rather simple implementation of a complex skill set.

>But you are not 'strengthing', you are 'rowing' or 'lifting' or
>'carrying'. You are doing some particular thing that builds strength.
>And 'knowing athletics' won't improve your Basketball skill any more
>than it will your Strength. That you must actually practice the skill
>to get the benefit is not relevant.

Then you happen to agree with me on this issue. The simple presence of the
athletics skill should not raise stats, like my interpretation of the
thread-originating post thought it sead.

Also, I'd like to point out that most roguelike characters don't
athleticise, either.

>
>>I really happen to like the skill tree idea -- stats up at the very top,
>>and skills make use of stats to modify their rolls, but the simple presence
>>or absence of a skill does not change the stat. This also prevents
>>unwanted codependency between independent stats, e.g. where your high skill
>>score im barbell lifting makes you better able to swing a sword [instead of
>>the strength gained from training barbell lifting improving your swing].
>
>No, you just want the strength gain from lifting barbells to
>miraculously improve my damage bonus when I kick. The only difference
>between the two cases you give above is that one adds a layer of
>indirection which the other does not. If the relationships are hidden
>from the player, he'll never see any difference.

So you don't mind physical stats at all, but instead think they're too
simplified? Strength should be for arms/legs/torso, Int should seperate
into memorization and logic, Cha should seperate into conversation and
leadership, and the like?

Root statistics, no matter how many there are [within reason], simplify
matters greatly. Without them, you have to keep track of N^2 relationships
between N skills that use X proficency. IE, if you have 7 weaponry skills,
barbell lifting, and elf-twirling, that's 56 two-way relationships to code
dependencies for. Instead, they could all improve arm-strength and take a
bonus from it.

>
>>IMO, a skill-only system has a problem where there's not enough mutual
>>improvement. Animal charming is not distinctly related to haggling, but
>>both through practice will improve your general charisma to some degree,
>>which will have an effect on the other.
>
>If you want that, it's easy enough to use a skill tree. But I disagree
>that these should cross-pollenate. Good with animals and good with
>people are quite different.

I never said they should cross-pollinate a lot. Being good with animals
will, however, impart some level of emotion-reading from facial expressions
and bahavior, which should definitely apply to humans.

>
>>Either the skill tree goes up to
>>one or several "root" skills, which are stats by another name, or it runs
>>the risk of eliminating the groupings of skills.
>
>I disagree that the top of a skill tree is a stat. Especially as games
>can and do have both skill trees and stats that are not part of the
>skill tree. Stats are generally treated quite differently from skills,
>whereas the top of the skill tree is just another skill that happens
>to not have another more general skill above it. And while stats
>pretend to be some basic fact about a character, maybe even a genetic
>one (although they're all too broad for that), the top of a skill tree
>represents experience and education.

Hm... I diverged from what I was thinking when I wrote that. The best way
to desribe myself would be an example:

Combat
Ranged Weaponry (uses Perception)
Bows
Composite Bow (Uses Strength)
[Note, this is rather simple. It computes only a single bonus. It does
not, and was not intended to, seperate damage and to-hit.]

The ultimate modifier, in this case, is the skill in Composite Bow + some
fraction of a Strength Bonus + the Bow skill + the Ranged Skill + some
fraction of PE Bonus + the Combat skill.

When used, experience is inherited up the tree -- some fraction goes
directly into Composite Bow, some goes into STR, the remainder goes to Bows
[which then divides it up between Bows and Ranged Weaponry, which then
divides it etc.]

With extended training with a composite bow, the character will get
stronger, have marginially better eyesight, and gain some general knowledge
of what to expect in combat. The strength and general combat knowledge
will transfer directly over when he decides to pick up a longsword. There
is no other connection between the composite bow and longsword, and there
doesn't need to be.

A skill tree with experience progression up the tree allows for addition of
skills without directly setting codependency rates.

>>I wasn't exactly clear on this. I was interpreting concentration strictly
>>as he wrote it, a skill whose sole function was to provide an MP bonus.
>
>So was I. I don't see why you think this can't be done. Again,
>Rolemaster was a very skill-based system. It began granting automatic
>spell point increases each level, but with Rolemaster Standard System,
>it moved to requiring developing spell points. It works. You may
>personally dislike it, but it is perfectly workable as a mechanic.

But if Concentration's only gameplay effect is to increase MP's, then how
on earth is a character supposed to practice it?

The comeback answer here is 'concentrate,' but I'll remind you that the
only gameplay effect of concentration is to increase MP's -- you have
concentration, you don't concentrate. The only way for an end-skill to be
practicable is to have some definite and direct gameplay effect.

>
>>The MP derived stat can be done entirely through Int/Wis/etc, and casting
>>costs can be discounted through proficencies in the individual spell, spell
>>college, and general magic (a magic skill tree, in essence).
>
>It certainly *can* be done entirely through stats, but this thread
>didn't start out with someone asking what can be done *without*
>skills, the question you seem interested in addressing, but in what
>can be done *with* them. The answer is, I think, almost anything you
>can imagine a use for.

Yes, but you could also have the concentration skill summon pink elephants
that tap-dance. It's just not logical.

The way he proposed concentration, it was a one-off skill that wasn't
really related to anything else. If there was any kind of skill grouping,
which it looked like there might be from the headers, then concentration
would overlap with any kind of generalized magic proficency, and IMO having
two skills that do one thing is something worth avoiding.

>
>You have a skill tree, a simple two level one with each 'type' of
>magic on the top and a set of individual spell skills underneath it.
>So, you have Enchantment skill and Runic Blade spell skill. Anyone can
>learn the Runic Blade spell and develop skill with it, but the skill
>check to see how much of a bonus is given by the Runic Blade
>enchantment is made against Enchantment skill + Runic Blade skill in a
>typical skill tree method. But that doesn't mean that Enchantment
>skill itself has to be trainable (the magic model may involve each

Under my currently-preferred skill tree model, the enchantment skill would
be trainable by use of enchantment spells. It doesn't have to be directly
trainable because it isn't end end-skill -- other skills branch off from
it.

>spell having a 'uniqueness' that prevent learning from one spell
>applying to another, which explains why the magic type skill is a
>fixed talent rather than a standard skill). An elf may have
>Enchantment 30 and a human only Enchantment 5 and the only way for a
>human to get that 30 is to change into an elf.

If it wasn't trainable [which essentially means that it's impossible to
gain any knowledge in it, IE that there is absolutely nothing in common
between Enchantment spells], then it's a talent, racial bonus, or whatever
you prefer to call it.


>>You may be right about summoning... summoning of mundane monsters could be
>>stuffed under the same spell genre as teleportation, as you're calling them
>>from somewhere and not making them from nothing.
>
>It isn't a point about summoning. It's that spells don't naturally
>fall into exclusive categories and don't work that well with rigid
>skill trees. You need to make some arbitrary decisions as to which
>tree to put it under, allow identical spells under multiple trees, or
>allow trees to 'cross' and have spells that belong in multiple trees
>and interact with multiple skills on the next level up. For my new,
>non-summoning example, a Steam Bolt spell -- fire and water magic
>both. *Really* makes your system tough if 'opposite' element magic
>types have some form of 'conflict' going on.

My skill tree is a system, not an environment. You do have another valid
point about steam bolt, however.

It could be solved by allowing multiple inheritence for skills -- in your
case, steam bolt would gain the average boni from Fire and Water schools of
magic. The system doesn't get broken unless a skill inherits a bonus from
another skill of equal or lower placement on the skill tree [IE, fire magic
inheriting a bonus from the frost bolt spell, or fireball inheriting a
bonus from fire bolt. Both a nono.]

>
>>Even with a mere 50 spells (and I'd love to see about a thousand or so.
>
>I'd like to see whichever is more appropriate to the world-vision of
>the game creator. The right 50 spells will go a long way and a
>thousand spells give you a thousand chances to have one that ruins
>game balance, so I can't go with a straight 'more is better'.

Of course, the wrong 50 spells can break a magic system entirely, making it
useless except for when it's vital.

>
>>:)), you could gain by creating fire/ice/air/earth elemental categories,
>>evocation, and divination. Evoocation handles the magic-missile, teleport,
>>and other pure-magic spells,

>supporting bits holding up ceiling over foe's head). Magic missile is


>directing raw magical energy and has more in common with enchanting or
>disenchanting an item (or better, with charging a wand), than with
>teleportation.

Depends on how wands are charged in your world. If they store raw magical
energy, sure, but it wouldn't work if they store refined magical energy one
step away from becoming the spell itself.

>
>>Woodcraft was badly named here... what would you call a Ranger's general
>>set of skills and proficencies? All of the ones I listed had to do with
>>outdoor, wilderness activities.
>
>I'd call it a set of skills that happen to be useful in the same place
>and are traditionally given to the stock game-fantasy character type
>called 'ranger'. And then I'd put the observation skills into my
>Observation skill tree and so on. I classify skills by what they are,
>not by which stereotype is most associated with them.

I'd rather classify them (at its root) by what mental processes you're
going through at the time [IE, stealth, pick pocket, and pick lock all tend
to be focused at least in part on avoiding detection], but whatever works.

>
>If you want to model familiarity with a particular environment, like
>the woods or cities or swamps, make that a skill. Your Rat-Man Skulker
>and your Wood Elf Ranger both have a nice high Stealth skill of 70,
>but the Skulker has an Urban Environment skill of 30 and the Ranger
>likewise has a Woodlands Environment skill of 30, so while each is

Not a bad idea.

>
>>True, although by most definitions of portals they're pretty simple things.
>>The bowling ball test works fine as an indicator, though.
>
>Not always. Sometimes the challenge is one a bowling ball wouldn't
>face, like a risk of insanity. It's *very* world-dependant to have a

Okay, okay, I concede the point. But I still think the idea of throwing a
bowling ball through a magical portal to be somewhat funny.

>

>>I'd argue that the magical quantum pixies are still directing magic energy
>>towards the goal of phase wall & teleport self, so they're still casting
>>the spell in some sense, albeit with an insane proficency/effectiveness.
>
>Not necessarily with an insane proficiency. Starting skill could be
>quite low. The difference between spell and skill is that they need
>not meet requirements for spell-casting (some games don't allow spell
>casting if blinded, silenced, confused, etc. or require some item,
>typically a book, in order to cast a spell) in order to use the skill
>and do not need spell points to use this natural ability.

The dividing part here would, IMO, be whether the movement-through-walls
requires some kind of unusual effort -- if not, it's a natural movement
ability.


>>I wasn't saying that it was only useful
>>for thief-type characters, I was saying it belonged in a group with the
>>other skills I called thiefy.
>
>I disagree that these make a natural grouping. Sneaking around is a
>'thiefly' art, but it has little to do with picking pockets. Nor does
>either have much to do with picking locks. You associate these skills
>because they are often useful to the same people, but that is a poor
>reason to put them in the same skill tree. If you want to mimic the
>training of traditional fantasy stereotypes, include some sort of
>Guild Training that let's you start with a set of relatively high
>starting skills appropriate to some stereotype.

Both skills, however, at least partially revolve around avoiding detection.
In such cases, you make efforts not to be seen, and to not be seen as
suspicious when the former's not an option. They all involve more
dexterity (manual dexterity or flexibility [to hide inside the barrel])
than brute strength, charisma, or memorization.

>Right, before you can do an appraisal, you need to identify the
>object. It's still too different processes and two different skills.
>And you don't need to know if it is *really* the fabled Ring of the
>Jeweler King in order to appraise it... as long as your mark can be
>convinced that it is.

Ah... different appraisal than angband/ADOM esque appraising, then.

>
>There's still not much you can do for injuries with herbs that you
>can't do without. Painkillers are pretty much it. And that's a matter
>of chewing on (or whatever) a prepared dose you got before hand, as
>trying to find a particular plant while in severe pain and suffering
>blood-loss is not likely to succeed no matter how skilled you are.

I think that there are a few mild natural antibiotics, if nothing else.
And who knows what else in a fantasy world. You're largely correct,
though.

>>No
>>amount of knowledge in your noggin is going to make that snakebite less
>>deadly if you don't do anything about it, and we already have a
>>healing/first-aid skill for treatment.
>
>That's the difference between having a skill and using it, nothing to
>do with skills vs. stats. The only thing that protects from snakebite

Well, as a general rule, skills are (or can become) verbs while Stats are
nouns. WRT a 'health' skill, use of it is called what... healthizing?
Helth is something you have, not something you do.

>
>Which is a learned skill. Actually, it's several learned skills. There
>are some apparently innate tendencies and potentials, but the closest
>thing I can think of to a general intelligence trait is curiosity --

At this point, roguelike development begins reaching the realm of
experimental psychology. :)

I think that neither of us are qualified to present an accurate model of a
character's intelligence, so any hashing out of an intelligence system
would depend greatly on the form we had in mind for other parts of the
hypothetical game.

R Dan Henry

unread,
Aug 27, 2001, 3:41:56 PM8/27/01
to
On Sat, 25 Aug 2001 15:42:19 GMT, the disembodied brain of

pfh...@gdi.net (Chris Subich) transmitted thus:

>On Sat, 25 Aug 2001 10:41:44 GMT, R Dan Henry <rdan...@earthlink.net>,
>the evil little half-monkey he is, wrote:

>That might work, but I still don't like the idea of a critical hits skill.

You are certainly entitled to personal preferences.

>In my mind, an end-level skill is something that either can be actively
>used or isn't terribly hard to practice. For example, Fire Magic makes an
>acceptable skill because you can go off and cast Fireball to your heart's
>content. Likewise, Sword Style (of fighting) makes sense as a skill
>because all you have to do to get better at it is go off and hack apart a
>few bears. How do you propose that a PC set about practicing his critical
>hits skill?

Ah, now you finally reveal a motive behind your preference. This is a
reasonable objection. And I cannot fully answer your question,
although I have to say that hacking at bears probably won't really
teach you all that much about killing armed and armored humanoids,
beyond basic weapon-handling (which can as well be learned in a
practice hall) and accustoming one to the sight of blood and stench of
death. Basically, one would practice Precision by practicing with
one's weapon... but with a different treatment of the target. Only the
points of vulnerability in the proper armor type or locations of vital
organs would count as hits. One practices not simply to score a hit,
but to score hits at a vulnerable location on the target. And this is
also a knowledge skill... one must learn details of anatomy and
structure of armor. Preferably by instruction, but if necessary, one
can always capture members of a species of interest for...
experimentation.

>If, as you say, critical hits are a function of fighting style, then the
>duelist is improving his proficency in that particular fighting style [of
>rapid, accurate hits], not just a random critical hit skill. Likewise, the
>barbarian improves his berzerker fighting style.

Yes, personally, I would prefer to use a form of styles rather than
have a skill for increasing critical hits. But such a skill is an
option.

>>>D&D has critical range based upon weapon type, and I happen to aggree with
>>>that -- perhaps allow existing combat boni to come into play [say the
>>>weapon has a critical hit ratio, and you criticially hit when your roll +
>>>boni is more than ratio times defender's roll + boni].
>>
>>Strange that you say 'D&D' has this. This is a crude version of
>>systems many games had 20 years ago.
>
>... and I would know this because of my excellent gameplaying skills when I
>wasn't yet in elementary school.

Which still makes it strange that you would phrase it this way.

>>>In my mind, skills should be something definite. In the case of combat,
>>>it's something you do -- swing a broadsword or dodge. A critical hit, to
>>>me, is a lucky or expert result from a hit -- you usually aren't aiming to
>>>do a critical hit over and above aiming to hit and cause damage to your
>>>enemy. A critical hit is a subset of a hit, so I don't think it needs its
>>>seperate skill.
>>
>>Well, swordfighting is a subset of fighting, so why have multiple
>>combat skills?
>
>Because there is (usually) a logical division between swordfighting and
>spear-fighting, and I can deliberately set out to practice either.

Okay, now that you talk about the question of how to practice a skill
I see what your problem with this skill is, but before you were
unclear as to what you meant. Nothing more to say here, I think.
Especially as we've become so long winded I'm not sure anyone is
paying attention at this point.

>>Yep. For hit point development. Of course, this doesn't even require
>>getting rid of stats. Rolemaster required you to spend development
>>points to increase concussion hits, but had a CON stat. But Rolemaster
>>is a real rpg, and getting rid of stats in a real rpg is difficult
>>because they give you something to work with in situations outside the
>>rules. With a roguelike, you don't need anything like that, because
>>everything that can happen must be explicitly programmed and a skill
>>can be assigned.
>
>Yes, but stats make an excellent foundation for skills. If I drink a
>potion of strength and get stronger, I'll hit harder, row faster, and break
>down more doors. The potion made me stronger -- it didn't provide a bonus
>to swordfighting, rowing, and door bashing.

Potions of strength exist largely as a means to exploit game
mechanics. And usually influence things in an illogical fashion. Sure,
magically boosting your strength would help you lift things and hit
harder, but it usually also increases skills that would logically go
*down* until you learned to adjust to your new strength. Control is
useful for everything except smashing stuff.

Mind you, I'm looking at stats for the theoretical original roguelike
I'll do 'someday', but I won't have an easy way to increase them. If
you're going to have stats to lay a foundation for skills, they ought
to stay (mostly) fixed. Otherwise, they aren't anymore fundamental as
aspects of your character and serve no purpose not better served by a
proper skill tree without stats.

>In implementation terms, a skill tree with stats used as modifiers makes
>for a rather simple implementation of a complex skill set.

Simplicity isn't necessary for a computer game, tho. The one advantage
of gaming on a computer is that it can do *lots* of number juggling.
GURPS needs stats because it needs defaults for its large number of
skills. Generating values for all those skills without such defaults
would be a royal pain when making characters by hand. However, in a
roguelike, you can generate all those skills painlessly (except for
the initial programming, of course, but you can make that as simple or
involved as you like depending on what factors you want to use...
race, culture, training, age, astrological sign, personality
traits,etc. or just 'set it to 10').

Stats aren't useless as game mechanisms, by any means, but they aren't
much on realism and they aren't indispensable.

>>But you are not 'strengthing', you are 'rowing' or 'lifting' or
>>'carrying'. You are doing some particular thing that builds strength.
>>And 'knowing athletics' won't improve your Basketball skill any more
>>than it will your Strength. That you must actually practice the skill
>>to get the benefit is not relevant.

>Then you happen to agree with me on this issue. The simple presence of the
>athletics skill should not raise stats, like my interpretation of the
>thread-originating post thought it sead.

The simple presence of athletics skills should not raise stats, no,
nor should it raise the skill itself, either. However, neither of
these was even proposed. What was proposed was that practice which
leads to increasing the skill should also increase the stat. Just as
any 'higher level' skill on a skill tree may be increased.

>Also, I'd like to point out that most roguelike characters don't
>athleticise, either.

Not sure about that. Not in Angband, as there is no way to do so, but
I think some of the others have training hall type places that would
logically include a gym. Nethackers push boulders around for exercise.
And they all get plenty of healthy walking and running, while carrying
large loads. They generally don't do 'athletic' things as such...
because they can't. You do have to include the possibility in the game
before the little @ can do it. Include the possibility and offer a
reward, and it will be done.

>So you don't mind physical stats at all, but instead think they're too
>simplified? Strength should be for arms/legs/torso, Int should seperate
>into memorization and logic, Cha should seperate into conversation and
>leadership, and the like?

No, I don't. I'd rather see stats eliminated than hugely multiplied.
Stats serve as a useful game mechanism for those who wish to use them.
To try to pretend to model reality with them is silly. The reality is
far too complex. Skills utilizes various combinations of muscle
groups, intellectual traits, personality quirks, etc. Trying to map
these is far too complex for the limited gain.

>Root statistics, no matter how many there are [within reason], simplify
>matters greatly. Without them, you have to keep track of N^2 relationships
>between N skills that use X proficency. IE, if you have 7 weaponry skills,
>barbell lifting, and elf-twirling, that's 56 two-way relationships to code
>dependencies for. Instead, they could all improve arm-strength and take a
>bonus from it.

Well, I wouldn't include a 'barbell lifting' skill, unless I were
doing a roguelike based on modern athletics. I would have a couple of
basic physical task skills like 'lifting', 'carrying', 'striking'.
'Striking' would be the one to give damage bonuses for weapons where a
strength bonus makes sense (no garrotes or whips or blowguns). Barbell
lifting would be modeled by going to the Gymnasium and spending time
and money for training on 'striking' and 'lifting' (you can develop
both with a good routine, for 'carrying' you need to work back and
legs mainly). And weapon skills already support each other if you have
even a simple skill tree. That leaves 'elf-twirling'. As elves are
magical creatures, I'd consider this a magical device skill and would
put it in the same tree with 'wand activation' and 'crystal control'
and certainly wouldn't let it help swordfighting.

>I never said they should cross-pollinate a lot. Being good with animals
>will, however, impart some level of emotion-reading from facial expressions
>and bahavior, which should definitely apply to humans.

Or misapply to humans. Sometimes the signals are similar, sometimes
almost opposite. Someone used to living among animals will find the
human custom of baring fangs to indicate happiness/friendliness odd.
Also, you don't even find consistency from one human culture to
another. Some things remain the same, but in other's your knowledge of
one culture will mislead you in the other. Of course, you can always
tie the two together at the top of the skill tree if you want.

I'd keep the two separate. I've known too many persons who can deal
well with humans and just have no clue as to animals. I'd also make a
third social category for 'spirits' if I had such a category, so
shamans didn't automatically become charming to humanoids.

>Combat
> Ranged Weaponry (uses Perception)
> Bows
> Composite Bow (Uses Strength)
>[Note, this is rather simple. It computes only a single bonus. It does
>not, and was not intended to, seperate damage and to-hit.]

>The ultimate modifier, in this case, is the skill in Composite Bow + some
>fraction of a Strength Bonus + the Bow skill + the Ranged Skill + some
>fraction of PE Bonus + the Combat skill.

Actually, I don't think this is a good model. If you want to represent
the value of strength in using a bow, give each bow a strength rating
required to use it properly and have a range of bows where stronger
bows have nicer bonuses. Trying to apply a great strength to a bow
meant for a lesser draw with just break the bow.

For statless, I'd set a certain requirement in skill, as bow-muscles
develop with practice (English/Welsh longbowmen supposed had
noticeably unequal arm sizes due to the exercise of archery.) And
height. Height is one stat that can't actually be replaced with
skills. Probably because it's one stat that actually represents
something concrete.

For melee, a 'striking' skill which includes not only strength
development, but general principles of applying that strength, is what
I'd use for damage bonuses if going statless. (How you apply your
strength matters, 'lifting' involves using proper technique just as
much as building muscle mass and 'carrying' involving getting a
well-balanced load.)

>With extended training with a composite bow, the character will get
>stronger, have marginially better eyesight, and gain some general knowledge
>of what to expect in combat. The strength and general combat knowledge
>will transfer directly over when he decides to pick up a longsword.

The motions of fighting with a longsword are quite different from
those of fighting with a bow. And combat for an archer is quite
different than melee. The transfer could actually be ignored
completely (although I don't recommend this, I *would* probably have
archery and swordfighting meet up higher in the skill tree) without
much loss of realism.

>There
>is no other connection between the composite bow and longsword, and there
>doesn't need to be.

You don't even need as much as that. Especially as it's not clear that
one actually much helps with the other, except as that in both cases,
if one advances the skill under actual battle conditions rather than
by training (which isn't very realistic), then one improves one's
poise in a crisis. One learns to ignore the screams of dying and focus
on killing and not being killed. But the actual skills have little in
common.

>A skill tree with experience progression up the tree allows for addition of
>skills without directly setting codependency rates.

I'm all for that. I'm just saying that one you have skill trees, you
don't *need* stats anymore. Not for a computer game. Possibly not even
for pen & paper roleplaying, but I have reservations there.

>But if Concentration's only gameplay effect is to increase MP's, then how
>on earth is a character supposed to practice it?

Depends on world-model. If magic points are 'gathered in' through some
form of meditation, practice would consist of meditating.

>The comeback answer here is 'concentrate,' but I'll remind you that the
>only gameplay effect of concentration is to increase MP's -- you have
>concentration, you don't concentrate. The only way for an end-skill to be
>practicable is to have some definite and direct gameplay effect.

Some skills may only be advanced by training and/or applying some
'general experience' points to it. Whether or not this is tolerable is
a matter of personal preference and game/world design decisions. It
might also be that you are practicing every time you regain your spell
points.

>>It certainly *can* be done entirely through stats, but this thread
>>didn't start out with someone asking what can be done *without*
>>skills, the question you seem interested in addressing, but in what
>>can be done *with* them. The answer is, I think, almost anything you
>>can imagine a use for.
>
>Yes, but you could also have the concentration skill summon pink elephants
>that tap-dance. It's just not logical.

>The way he proposed concentration, it was a one-off skill that wasn't
>really related to anything else. If there was any kind of skill grouping,
>which it looked like there might be from the headers, then concentration
>would overlap with any kind of generalized magic proficency, and IMO having
>two skills that do one thing is something worth avoiding.

Well, the initial skill groupings were terrible, so I can see how they
could cause problems. I also agree that duplicating skills is
*usually* to be avoided. However, he wasn't proposing a system, just
throwing out ideas, so we need not assume all the skills listed would
be used in the same game. You really need to think up what you want
'magic' to be in your world before you figure out what skills to use.

>>You have a skill tree, a simple two level one with each 'type' of
>>magic on the top and a set of individual spell skills underneath it.
>>So, you have Enchantment skill and Runic Blade spell skill. Anyone can
>>learn the Runic Blade spell and develop skill with it, but the skill
>>check to see how much of a bonus is given by the Runic Blade
>>enchantment is made against Enchantment skill + Runic Blade skill in a
>>typical skill tree method. But that doesn't mean that Enchantment
>>skill itself has to be trainable (the magic model may involve each

>Under my currently-preferred skill tree model, the enchantment skill would
>be trainable by use of enchantment spells. It doesn't have to be directly
>trainable because it isn't end end-skill -- other skills branch off from
>it.

I understand that. I'm trying to explain to you why a designer might
not want this skill to be trainable *at all*.

>>spell having a 'uniqueness' that prevent learning from one spell
>>applying to another, which explains why the magic type skill is a
>>fixed talent rather than a standard skill). An elf may have
>>Enchantment 30 and a human only Enchantment 5 and the only way for a
>>human to get that 30 is to change into an elf.
>
>If it wasn't trainable [which essentially means that it's impossible to
>gain any knowledge in it, IE that there is absolutely nothing in common
>between Enchantment spells], then it's a talent, racial bonus, or whatever
>you prefer to call it.

You can call it that if you like, however, if it otherwise functions
like any other head of skill tree, I find it tolerable to call it a
skill. However, it certainly represent the level of skill, which is
all I said to begin with. If you want to call it a 'talent', I think
that's perfectly fine terminology. But then recognize that your
approach doesn't work if the designer prefers to use talents for spell
groups rather than skills.

>>>You may be right about summoning... summoning of mundane monsters could be
>>>stuffed under the same spell genre as teleportation, as you're calling them
>>>from somewhere and not making them from nothing.
>>
>>It isn't a point about summoning. It's that spells don't naturally
>>fall into exclusive categories and don't work that well with rigid
>>skill trees. You need to make some arbitrary decisions as to which
>>tree to put it under, allow identical spells under multiple trees, or
>>allow trees to 'cross' and have spells that belong in multiple trees
>>and interact with multiple skills on the next level up. For my new,
>>non-summoning example, a Steam Bolt spell -- fire and water magic
>>both. *Really* makes your system tough if 'opposite' element magic
>>types have some form of 'conflict' going on.
>
>My skill tree is a system, not an environment. You do have another valid
>point about steam bolt, however.
>
>It could be solved by allowing multiple inheritence for skills -- in your
>case, steam bolt would gain the average boni from Fire and Water schools of
>magic. The system doesn't get broken unless a skill inherits a bonus from
>another skill of equal or lower placement on the skill tree [IE, fire magic
>inheriting a bonus from the frost bolt spell, or fireball inheriting a
>bonus from fire bolt. Both a nono.]

Yes, you can deal with it that way. Or in several other ways. I never
said the problems were insurmountable. I said "your approach creates
complication in multiple type spells" -- not that it can't work.
Crossing skill trees, however, brings up its own questions. Once you
allow that multiple inheritance is possible, you'll hear all sorts of
arguments for why it should apply to other skills. It's a respectable
approach, but I still say that if you keep your spell list short,
grouping them is unnecessary.

>>>:)), you could gain by creating fire/ice/air/earth elemental categories,
>>>evocation, and divination. Evoocation handles the magic-missile, teleport,
>>>and other pure-magic spells,
>
>>supporting bits holding up ceiling over foe's head). Magic missile is
>>directing raw magical energy and has more in common with enchanting or
>>disenchanting an item (or better, with charging a wand), than with
>>teleportation.
>
>Depends on how wands are charged in your world. If they store raw magical
>energy, sure, but it wouldn't work if they store refined magical energy one
>step away from becoming the spell itself.

Depends on how refined you view the energy in magic missile, I guess.
But even if you think that the skill for the spell in the wand is
needed to *shape* the energy before placing it in the wand, putting it
into the wand is still a mana-channeling process and similar to
sending mana into a living body. If I used your second model, I'd just
require two skill checks.

>>I'd call it a set of skills that happen to be useful in the same place
>>and are traditionally given to the stock game-fantasy character type
>>called 'ranger'. And then I'd put the observation skills into my
>>Observation skill tree and so on. I classify skills by what they are,
>>not by which stereotype is most associated with them.
>
>I'd rather classify them (at its root) by what mental processes you're
>going through at the time [IE, stealth, pick pocket, and pick lock all tend
>to be focused at least in part on avoiding detection], but whatever works.

Stealth, yes. Picking pockets, yes. Although typically in different
way. If you are picking a lock on a door with a group of orcs down the
corridor, you will be worried about being stealthy about it. If you
are trying to unlock a treasure chest you've brought back home with
you, you really aren't worried about it. And I regard it as the same
skill in both cases. Your classification scheme would make any decent
outdoorsman (who can move stealthily) into a natural at thieving. That
just doesn't make sense to me.

>>If you want to model familiarity with a particular environment, like
>>the woods or cities or swamps, make that a skill. Your Rat-Man Skulker
>>and your Wood Elf Ranger both have a nice high Stealth skill of 70,
>>but the Skulker has an Urban Environment skill of 30 and the Ranger
>>likewise has a Woodlands Environment skill of 30, so while each is
>
>Not a bad idea.

Thanks. I know I've seen it before for combat, I'm not sure if I came
up with applying it more widely on my own or saw it suggested in some
forum. It's probably not original with me, tho.

>>>I'd argue that the magical quantum pixies are still directing magic energy
>>>towards the goal of phase wall & teleport self, so they're still casting
>>>the spell in some sense, albeit with an insane proficency/effectiveness.
>>
>>Not necessarily with an insane proficiency. Starting skill could be
>>quite low. The difference between spell and skill is that they need
>>not meet requirements for spell-casting (some games don't allow spell
>>casting if blinded, silenced, confused, etc. or require some item,
>>typically a book, in order to cast a spell) in order to use the skill
>>and do not need spell points to use this natural ability.
>
>The dividing part here would, IMO, be whether the movement-through-walls
>requires some kind of unusual effort -- if not, it's a natural movement
>ability.

Climbing a tree requires an unusual effort, as does climbing a wall,
or swimming against a current. At least, these all require skill
checks, which is the degree of 'unusual effort' a quantum pixie would
need to move through walls. Yet these are all natural movement
abilities.

>>I disagree that these make a natural grouping. Sneaking around is a
>>'thiefly' art, but it has little to do with picking pockets. Nor does
>>either have much to do with picking locks. You associate these skills
>>because they are often useful to the same people, but that is a poor
>>reason to put them in the same skill tree. If you want to mimic the
>>training of traditional fantasy stereotypes, include some sort of
>>Guild Training that let's you start with a set of relatively high
>>starting skills appropriate to some stereotype.
>
>Both skills, however, at least partially revolve around avoiding detection.

Yes, but they involve avoiding detection in quite different ways.
Sneaking around means hiding *yourself*, picking pockets means hiding
a *particular activity*. If you acted like someone sneaking around
while trying to pick pockets, you'd be detected more readily.

>In such cases, you make efforts not to be seen, and to not be seen as
>suspicious when the former's not an option. They all involve more
>dexterity (manual dexterity or flexibility [to hide inside the barrel])
>than brute strength, charisma, or memorization.

Ah, so this is really driven by your stat-based thinking! Actually,
picking pockets without being caught *ought* to be based on charisma
as well as dexterity under the D&D stat scheme. A lot of it is
understanding people -- recognizing when they are distracted, keeping
an innocent look on your face, etc.

And my homebrewed stat schemes have *always* split general body
agility (the one for stealth) and manual dexterity (the one that helps
pick pockets). So I've never seen them alike on that point.

And I'm not sure that high agility is that useful for stealth...
unless you are trying to move quickly at the same time. I'm not at all
agile, but I sneak up on people without even trying.

>>Right, before you can do an appraisal, you need to identify the
>>object. It's still too different processes and two different skills.
>>And you don't need to know if it is *really* the fabled Ring of the
>>Jeweler King in order to appraise it... as long as your mark can be
>>convinced that it is.

>Ah... different appraisal than angband/ADOM esque appraising, then.

Angband doesn't have 'appraisal'. It has pseudo-ID, ID, & *ID*. But,
yes, I'm talking real-world meaning here... what can I get for this?

>>There's still not much you can do for injuries with herbs that you
>>can't do without. Painkillers are pretty much it. And that's a matter
>>of chewing on (or whatever) a prepared dose you got before hand, as
>>trying to find a particular plant while in severe pain and suffering
>>blood-loss is not likely to succeed no matter how skilled you are.
>
>I think that there are a few mild natural antibiotics, if nothing else.

Cleaning and bandaging your wounds properly is far more important,
tho. Washing your clothes regularly might help more than anything, as
they were often the source of contamination.

>And who knows what else in a fantasy world. You're largely correct,
>though.

Certainly in a fantasy world you can have a magical herb garden and
skills related to it would be useful. But I don't see them replacing
field medicine as a skill.

>Well, as a general rule, skills are (or can become) verbs while Stats are
>nouns. WRT a 'health' skill, use of it is called what... healthizing?
>Helth is something you have, not something you do.

So the name can be changed... Recovery/Recovering. How's that?

>>Which is a learned skill. Actually, it's several learned skills. There
>>are some apparently innate tendencies and potentials, but the closest
>>thing I can think of to a general intelligence trait is curiosity --
>
>At this point, roguelike development begins reaching the realm of
>experimental psychology. :)
>
>I think that neither of us are qualified to present an accurate model of a
>character's intelligence, so any hashing out of an intelligence system
>would depend greatly on the form we had in mind for other parts of the
>hypothetical game.

It's too early for anyone to present an accurate model of what is real
in 'intelligence'. It is far too late to pretend that the D&D model
has any meaningful relationship to the real world. That's one reason I
support a statless system. Skills are generally sufficiently
well-defined to get a handle on.

Chris Subich

unread,
Aug 27, 2001, 9:24:38 PM8/27/01
to
On Mon, 27 Aug 2001 19:41:56 GMT, R Dan Henry <rdan...@earthlink.net>,
the evil little half-monkey he is, wrote:

>On Sat, 25 Aug 2001 15:42:19 GMT, the disembodied brain of
>pfh...@gdi.net (Chris Subich) transmitted thus:
>

>>In my mind, an end-level skill is something that either can be actively
>>used or isn't terribly hard to practice. For example, Fire Magic makes an
>>acceptable skill because you can go off and cast Fireball to your heart's
>>content. Likewise, Sword Style (of fighting) makes sense as a skill
>>because all you have to do to get better at it is go off and hack apart a
>>few bears. How do you propose that a PC set about practicing his critical
>>hits skill?
>
>Ah, now you finally reveal a motive behind your preference. This is a
>reasonable objection. And I cannot fully answer your question,

I was attempting to imply it earlier, but I think we were using slightly
different criteria for what constitutes a skill. My fault.

>although I have to say that hacking at bears probably won't really
>teach you all that much about killing armed and armored humanoids,

>but to score hits at a vulnerable location on the target. And this is
>also a knowledge skill... one must learn details of anatomy and
>structure of armor. Preferably by instruction, but if necessary, one
>can always capture members of a species of interest for...
>experimentation.

This would be interesting to see put in... it could be done fairly nicely,
IMO, with a general "knowledge" data field per type of monster (much like
ADOM's Monster Memory does for all of the monster's abilities) - use of
that knolwedge field [which could be incrememted through means besides
combat] as a bonus to combat/critical hits would improve that combat.

Maybe, as has been suggested before, a monster "tree" for general knowledge
increases.


>>In implementation terms, a skill tree with stats used as modifiers makes
>>for a rather simple implementation of a complex skill set.
>
>Simplicity isn't necessary for a computer game, tho. The one advantage
>of gaming on a computer is that it can do *lots* of number juggling.

True, but the system either has to be simple or fairly intuitive/logical if
you expect your players to completely grasp it.

>>Root statistics, no matter how many there are [within reason], simplify
>>matters greatly. Without them, you have to keep track of N^2 relationships
>>between N skills that use X proficency. IE, if you have 7 weaponry skills,
>>barbell lifting, and elf-twirling, that's 56 two-way relationships to code
>>dependencies for. Instead, they could all improve arm-strength and take a
>>bonus from it.
>

>even a simple skill tree. That leaves 'elf-twirling'. As elves are
>magical creatures, I'd consider this a magical device skill and would
>put it in the same tree with 'wand activation' and 'crystal control'
>and certainly wouldn't let it help swordfighting.

Lol.

<snip a bunch of detail and stuff that we're not too far apart on>


>>A skill tree with experience progression up the tree allows for addition of
>>skills without directly setting codependency rates.
>
>I'm all for that. I'm just saying that one you have skill trees, you
>don't *need* stats anymore. Not for a computer game. Possibly not even
>for pen & paper roleplaying, but I have reservations there.

This is a really interesting topic worthy of its own discussion.

I'm still not quite clear how you will model physical attributes in a
skill-only system -- is it all hidden directly in the skills, or is there
some upper-level "root" for the skills (and/or display)?

>Well, the initial skill groupings were terrible, so I can see how they
>could cause problems. I also agree that duplicating skills is
>*usually* to be avoided. However, he wasn't proposing a system, just
>throwing out ideas, so we need not assume all the skills listed would
>be used in the same game. You really need to think up what you want
>'magic' to be in your world before you figure out what skills to use.

True. Part of the problem was that it's difficult to simply throw out
skills without a semi-established system to provide framework for it.


>>>I'd call it a set of skills that happen to be useful in the same place
>>>and are traditionally given to the stock game-fantasy character type
>>>called 'ranger'. And then I'd put the observation skills into my
>>>Observation skill tree and so on. I classify skills by what they are,
>>>not by which stereotype is most associated with them.
>>
>>I'd rather classify them (at its root) by what mental processes you're
>>going through at the time [IE, stealth, pick pocket, and pick lock all tend
>>to be focused at least in part on avoiding detection], but whatever works.
>
>Stealth, yes. Picking pockets, yes. Although typically in different
>way. If you are picking a lock on a door with a group of orcs down the
>corridor, you will be worried about being stealthy about it. If you
>are trying to unlock a treasure chest you've brought back home with
>you, you really aren't worried about it. And I regard it as the same
>skill in both cases. Your classification scheme would make any decent
>outdoorsman (who can move stealthily) into a natural at thieving. That
>just doesn't make sense to me.
>

I never said they'd be closely related. What set of skills would you stick
pick lock with?

>>>>I'd argue that the magical quantum pixies are still directing magic energy
>>>>towards the goal of phase wall & teleport self, so they're still casting
>>>>the spell in some sense, albeit with an insane proficency/effectiveness.
>>>
>>>Not necessarily with an insane proficiency. Starting skill could be
>>>quite low. The difference between spell and skill is that they need
>>>not meet requirements for spell-casting (some games don't allow spell
>>>casting if blinded, silenced, confused, etc. or require some item,
>>>typically a book, in order to cast a spell) in order to use the skill
>>>and do not need spell points to use this natural ability.
>>
>>The dividing part here would, IMO, be whether the movement-through-walls
>>requires some kind of unusual effort -- if not, it's a natural movement
>>ability.
>
>Climbing a tree requires an unusual effort, as does climbing a wall,
>or swimming against a current. At least, these all require skill
>checks, which is the degree of 'unusual effort' a quantum pixie would
>need to move through walls. Yet these are all natural movement
>abilities.

But a person must be trained to climb a tree/wall/swim -- that's why it's a
skill. A cat, however, doesn't, and thus climbing [at least] is a natural
movement ability. The question is whether a quanatumn pixie has to
learn/be taught how to fly through walls or can do it naturally.

>And I'm not sure that high agility is that useful for stealth...
>unless you are trying to move quickly at the same time. I'm not at all
>agile, but I sneak up on people without even trying.

Well, I was thinking more in terms of flexibility -- being able to fold
yourself in half to fit into the chest is a very nice way to hide.

>>Ah... different appraisal than angband/ADOM esque appraising, then.
>
>Angband doesn't have 'appraisal'. It has pseudo-ID, ID, & *ID*. But,
>yes, I'm talking real-world meaning here... what can I get for this?

You can tell how much angband I regularly play. :)

R Dan Henry

unread,
Aug 28, 2001, 12:38:31 AM8/28/01
to
On Tue, 28 Aug 2001 01:24:38 GMT, the disembodied brain of

pfh...@gdi.net (Chris Subich) transmitted thus:

>On Mon, 27 Aug 2001 19:41:56 GMT, R Dan Henry <rdan...@earthlink.net>,
>the evil little half-monkey he is, wrote:

>>On Sat, 25 Aug 2001 15:42:19 GMT, the disembodied brain of
>>pfh...@gdi.net (Chris Subich) transmitted thus:

>I was attempting to imply it earlier, but I think we were using slightly


>different criteria for what constitutes a skill. My fault.

Well, a lot of any typical detailed discussion gets down to digging up
presuppositions on both sides. I have to say that this has been a
valuable conversation for me in that its made me think over some
things so that I'm now thinking that if I ever do get around to doing
an original roguelike (I'm currently just trying to create an Angband
variant), that I'll at least consider making it statless. I've now
thought of ways to do this that hadn't occurred to me before.

>>but to score hits at a vulnerable location on the target. And this is
>>also a knowledge skill... one must learn details of anatomy and
>>structure of armor. Preferably by instruction, but if necessary, one
>>can always capture members of a species of interest for...
>>experimentation.

>This would be interesting to see put in... it could be done fairly nicely,
>IMO, with a general "knowledge" data field per type of monster (much like
>ADOM's Monster Memory does for all of the monster's abilities) - use of
>that knolwedge field [which could be incrememted through means besides
>combat] as a bonus to combat/critical hits would improve that combat.

I'd already planned for a lot of 'knowledge' skills. Of course, my
hope is to someday make a sci-fi roguelike, so it's natural to have
knowledge skills aplenty there. The proper (xeno)biological skill
could be consulted to modify critical hit chances, although I'm not
sure how to model this when aliens attack the hero without doing full
skill sets for each of them, which may well be more data than I want
to be shoveling around. Oh, well, could just do skill templates for
each race and profession and have all rookie Xoznark guard have the
same skills across the board. While unrealistic, you probably wouldn't
notice they all have the same Gambling with Yongrag Sticks skill
anyway.

>Maybe, as has been suggested before, a monster "tree" for general knowledge
>increases.

I'd thought of something like:
Biology----(given world)----(species of world). For a fantasy
world, I'd use the middle level for
mammal/reptile/avian/plant/fungus/etc.

>>>In implementation terms, a skill tree with stats used as modifiers makes
>>>for a rather simple implementation of a complex skill set.

>>Simplicity isn't necessary for a computer game, tho. The one advantage
>>of gaming on a computer is that it can do *lots* of number juggling.

>True, but the system either has to be simple or fairly intuitive/logical if
>you expect your players to completely grasp it.

The player doesn't need to *completely* grasp it. To what degree the
player need to understand the mechanics of the system varies widely
depending on the game. Angband's mainly a tactical exercise and it
can't really hide too much information from the player without
limiting his ability to play the game well. Conversely, spoilers
really don't spoil Angband. Other roguelike are more
'world-exploratory', being very playable without a focus on winning,
just on seeing a little more and learning a little more. These can
conceal much more and also lose much more if you read spoilers. The
main thing that matters is that the player can understand what the
information he does have access to means. If he knows an 88 Skill in
Mango-Tossing means he can hit the Greater Clown of Evil about half
the time, he doesn't need to know *all* the factors that might go into
what may be a very complex determination of the skill value. Just let
him know things like Mango-Tossing skill is benefited by increases in
the higher level Fruit-Throwing skill or wearing Gloves of Hurling.

What shouldn't happen is misleading things. If your bonuses work in a
non-linear fashion, so that a Knife (+10) may not give you twice the
bonus of a Knife (+5), don't display numbers. Call them Steel Knife
and Iron Knife instead and just offer a list showing the order of the
value of metals (or make that available within the game).

>>>A skill tree with experience progression up the tree allows for addition of
>>>skills without directly setting codependency rates.
>>
>>I'm all for that. I'm just saying that one you have skill trees, you
>>don't *need* stats anymore. Not for a computer game. Possibly not even
>>for pen & paper roleplaying, but I have reservations there.
>
>This is a really interesting topic worthy of its own discussion.

>I'm still not quite clear how you will model physical attributes in a
>skill-only system -- is it all hidden directly in the skills, or is there
>some upper-level "root" for the skills (and/or display)?

If your character can do a lot of strong-guy stuff, you can call him
strong. There will be some relationships between many of these skills
as they'll be in the same skill trees. Attributes like height, weight,
and eye color are handled normally. But the game won't give you a
value for traditional stats, just skill levels. Some of these are
going to be fairly general. If you look at 'lifting', 'carrying', and
'striking' skills, you can justly call someone with high scores in all
of these strong, as someone who in the real world did lots of lifting,
carrying, and striking would be strong. But a low skill might just
indicate a poor application of existing muscle strength, so it isn't
just a split-up stat. One reason I was thinking of stats in the
beginning was to help distinguish alien species without loading them
down with exotic (and implausible) abilities -- not that I'm against
exotic abilities in a game, it just seems overkill to make them
common. But I think this approach will let me make racial skill
templates that will do an even better job.

For a fantasy example, consider the centaur. Now if his human torso is
roughly the same size as that of a man, he won't be naturally better
at striking things, as his arms are the same as a man's. But if he
learns to lift properly, with his legs, he's got four of those, so he
should get a bonus. But not a huge bonus as the human arms and back
still have to do a fair portion of the work. But carrying! That mighty
equine portion is most beneficial to this task. A centaur's carrying
skill bonus would be very nice.

>I never said they'd be closely related. What set of skills would you stick
>pick lock with?

Off the top of my head, without having actually worked out a set of
skill trees, I'd say probably something like this:

Tool Manipulation----Large Tools----Smithing----Etc.
\_Fine Tools----Engraving
\_Jeweler
|_Picking Locks
|_Making Locks

>>>>>I'd argue that the magical quantum pixies are still directing magic energy
>>>>>towards the goal of phase wall & teleport self, so they're still casting
>>>>>the spell in some sense, albeit with an insane proficency/effectiveness.
>>>>
>>>>Not necessarily with an insane proficiency. Starting skill could be
>>>>quite low. The difference between spell and skill is that they need
>>>>not meet requirements for spell-casting (some games don't allow spell
>>>>casting if blinded, silenced, confused, etc. or require some item,
>>>>typically a book, in order to cast a spell) in order to use the skill
>>>>and do not need spell points to use this natural ability.
>>>
>>>The dividing part here would, IMO, be whether the movement-through-walls
>>>requires some kind of unusual effort -- if not, it's a natural movement
>>>ability.
>>
>>Climbing a tree requires an unusual effort, as does climbing a wall,
>>or swimming against a current. At least, these all require skill
>>checks, which is the degree of 'unusual effort' a quantum pixie would
>>need to move through walls. Yet these are all natural movement
>>abilities.
>
>But a person must be trained to climb a tree/wall/swim -- that's why it's a
>skill. A cat, however, doesn't, and thus climbing [at least] is a natural
>movement ability. The question is whether a quanatumn pixie has to
>learn/be taught how to fly through walls or can do it naturally.

It is a learned ability, although one for which they have a naturally
high aptitude as well as the necessary physical/magical constitution.
It isn't quite as easy as walking, more like climbing a tree. But
remember, you had to learn even to walk. Walking is a skill, just one
that we have developed to such a high degree we almost never fail a
skill check unless there is some large modifier:

Extremely Pretty Girl in Short Skirt -60 to all rolls.

>>And I'm not sure that high agility is that useful for stealth...
>>unless you are trying to move quickly at the same time. I'm not at all
>>agile, but I sneak up on people without even trying.

>Well, I was thinking more in terms of flexibility -- being able to fold
>yourself in half to fit into the chest is a very nice way to hide.

I think I've seen a system that put Stealth under Concentration
skills. The ability to be careful seems more important than either
agility or flexibility, although good balance is very helpful when
freezing midstride. The situation you describe above would seem to me
to involve use of a separate Contortionist skill rather than being
straight hiding. Success at that skill then gives a large bonus to
Hiding. Unless somebody tries to get something out of the chest.

Anyway the real flexibility comes when you try to get out again! :-)

Robot Monster

unread,
Aug 29, 2001, 1:02:16 AM8/29/01
to
R Dan Henry <rdan...@earthlink.net> communicated:

>
> On Tue, 28 Aug 2001 01:24:38 GMT, the disembodied brain of
> pfh...@gdi.net (Chris Subich) transmitted thus:
> >
> >I was attempting to imply it earlier, but I think we were using slightly
> >different criteria for what constitutes a skill. My fault.
>
> Well, a lot of any typical detailed discussion gets down to digging up
> presuppositions on both sides. I have to say that this has been a
> valuable conversation for me in that its made me think over some
> things so that I'm now thinking that if I ever do get around to doing
> an original roguelike (I'm currently just trying to create an Angband
> variant), that I'll at least consider making it statless. I've now
> thought of ways to do this that hadn't occurred to me before.
>

[Anatomy knowledge affecting critical hits]



> >This would be interesting to see put in... it could be done fairly nicely,
> >IMO, with a general "knowledge" data field per type of monster (much like
> >ADOM's Monster Memory does for all of the monster's abilities) - use of
> >that knolwedge field [which could be incrememted through means besides
> >combat] as a bonus to combat/critical hits would improve that combat.
>
> I'd already planned for a lot of 'knowledge' skills. Of course, my
> hope is to someday make a sci-fi roguelike, so it's natural to have
> knowledge skills aplenty there. The proper (xeno)biological skill
> could be consulted to modify critical hit chances, although I'm not
> sure how to model this when aliens attack the hero without doing full
> skill sets for each of them, which may well be more data than I want
> to be shoveling around.

You could use several knowledge-type skills for this, such as:

Knowledge: ___ Anatomy: ___ Xoznark
| |_ Urrlak
| \_ Zelion
|
|_ Culture ___ Xoznark
| |_ Urrlak
| \_ Zelion
|
\_ Communication ___ Xoznark
|_ Urrlak
\_ Zelion

and then use these skills as modifiers when dealing with creatures of
other species (in the same way as the proposed environment-advantage
skills).

> Oh, well, could just do skill templates for
> each race and profession and have all rookie Xoznark guard have the
> same skills across the board. While unrealistic, you probably wouldn't
> notice they all have the same Gambling with Yongrag Sticks skill
> anyway.

Is Yongrag Stick a race or a form of gambling? Or both? I'm
picturing a version of cock-fighting using large stick insects.

>
> >Maybe, as has been suggested before, a monster "tree" for general knowledge
> >increases.
>
> I'd thought of something like:
> Biology----(given world)----(species of world). For a fantasy
> world, I'd use the middle level for
> mammal/reptile/avian/plant/fungus/etc.
>

I'd not go into much detail further down the tree. It would take an
awful lot of storage to retain data relating to anatomy knowledge for
each individual animal and you wouldn't need more than three simple
levels, eg. anatomy->reptile->lizard; anatomy->reptile->snake. You
have to ask, 'How different are these animals from each other?' and,
'What makes them different?' In a lot of cases there will only be
major anatomical differences between animals of different higher-level
categories, eg. lizard and mammal. Unless you intend on making a
doctor/veterinarian class who can specialise in healing particular
animals, this level of detail can be kept simple.
If you go down then to define skills such as Newt anatomy and Iguana
anatomy you might upset players who love detail if you don't define
the skills ToolManipulation->Fine Tools->Picking Locks->FiveTumblerKey
and ToolManipulation->Fine Tools->Picking
Locks->ThreeTumblerCombination : it is quite realistic to have an
expert at one of these skills who is hopeless at the other. (But of
course combination locks would be under perception anyway...)

> >>>In implementation terms, a skill tree with stats used as modifiers makes
> >>>for a rather simple implementation of a complex skill set.
>
> >>Simplicity isn't necessary for a computer game, tho. The one advantage
> >>of gaming on a computer is that it can do *lots* of number juggling.
>
> >True, but the system either has to be simple or fairly intuitive/logical if
> >you expect your players to completely grasp it.
>
> The player doesn't need to *completely* grasp it. To what degree the
> player need to understand the mechanics of the system varies widely
> depending on the game. Angband's mainly a tactical exercise and it
> can't really hide too much information from the player without
> limiting his ability to play the game well. Conversely, spoilers
> really don't spoil Angband. Other roguelike are more
> 'world-exploratory', being very playable without a focus on winning,
> just on seeing a little more and learning a little more. These can
> conceal much more and also lose much more if you read spoilers. The
> main thing that matters is that the player can understand what the
> information he does have access to means. If he knows an 88 Skill in
> Mango-Tossing means he can hit the Greater Clown of Evil about half
> the time, he doesn't need to know *all* the factors that might go into
> what may be a very complex determination of the skill value. Just let
> him know things like Mango-Tossing skill is benefited by increases in
> the higher level Fruit-Throwing skill or wearing Gloves of Hurling.

This really is how I felt when I first played Nethack - I'd been
playing Angband for a good three years before I was inroduced to the
wider world of roguelikes. Nethack was a blessing because of the
simplicity of its item enchantments: good or bad. No dice rolls for
damage are displayed. For the most part, common sense should be
relied upon for working out weapon preferences. Sharp or blunt (in
the case of edged weapons) is all the player really need know, and I'd
be willing to say they don't even need to know *exactly* how good the
item is. Many players would be satisfied with the description, 'a
fine, sharp elven dagger' for an elven dagger with reasonable plusses
to hit (fine) and to damage (sharp).


> What shouldn't happen is misleading things. If your bonuses work in a
> non-linear fashion, so that a Knife (+10) may not give you twice the
> bonus of a Knife (+5), don't display numbers. Call them Steel Knife
> and Iron Knife instead and just offer a list showing the order of the
> value of metals (or make that available within the game).

Of course using alloys as weapon prefixes would mean renaming scrolls
of enchant weapon to scrolls of upgrade material. Unless you pulled
the console RPG trick and simply set weapon quality as fixed (ie. the
only way to get a better weapon than a ceramic sword is to ditch it
and buy a titanium sword or (better yet!) a Laconian sword.

> >>>A skill tree with experience progression up the tree allows for addition of
> >>>skills without directly setting codependency rates.
> >>
> >>I'm all for that. I'm just saying that one you have skill trees, you
> >>don't *need* stats anymore. Not for a computer game. Possibly not even
> >>for pen & paper roleplaying, but I have reservations there.
> >
> >This is a really interesting topic worthy of its own discussion.
>
> >I'm still not quite clear how you will model physical attributes in a
> >skill-only system -- is it all hidden directly in the skills, or is there
> >some upper-level "root" for the skills (and/or display)?
>
> If your character can do a lot of strong-guy stuff, you can call him
> strong. There will be some relationships between many of these skills
> as they'll be in the same skill trees. Attributes like height, weight,
> and eye color are handled normally. But the game won't give you a
> value for traditional stats, just skill levels. Some of these are
> going to be fairly general. If you look at 'lifting', 'carrying', and
> 'striking' skills, you can justly call someone with high scores in all
> of these strong, as someone who in the real world did lots of lifting,
> carrying, and striking would be strong. But a low skill might just
> indicate a poor application of existing muscle strength, so it isn't
> just a split-up stat.

If there is any way in the game of increasing/decreasing muscle
strength other than through exercise (ie. through magic) then you
should have some way of mapping this to your skill set. You note that
skills like lifting are not just a split-up of ability and technique.
Could two characters exist who could lift the same weight where char1
did it using brute strength and char2 did it using a combination of
strength and lifting techniques? If so, would a potion of strength
give more of a boost to the first character, because the second relies
less on strength? Would a potion of Forget Climbing Ability affect
them equally?

> One reason I was thinking of stats in the
> beginning was to help distinguish alien species without loading them
> down with exotic (and implausible) abilities -- not that I'm against
> exotic abilities in a game, it just seems overkill to make them
> common. But I think this approach will let me make racial skill
> templates that will do an even better job.
>
> For a fantasy example, consider the centaur. Now if his human torso is
> roughly the same size as that of a man, he won't be naturally better
> at striking things, as his arms are the same as a man's. But if he
> learns to lift properly, with his legs, he's got four of those, so he
> should get a bonus. But not a huge bonus as the human arms and back
> still have to do a fair portion of the work. But carrying! That mighty
> equine portion is most beneficial to this task. A centaur's carrying
> skill bonus would be very nice.
>
> >I never said they'd be closely related. What set of skills would you stick
> >pick lock with?
>
> Off the top of my head, without having actually worked out a set of
> skill trees, I'd say probably something like this:
>
> Tool Manipulation----Large Tools----Smithing----Etc.
> \_Fine Tools----Engraving
> \_Jeweler
> |_Picking Locks
> |_Making Locks
>

The problem with skill trees rears its ugly head. Jeweler might be
better off being a Art/Craft skill with a bonus given from Fine Tools
skill.

> >>And I'm not sure that high agility is that useful for stealth...
> >>unless you are trying to move quickly at the same time. I'm not at all
> >>agile, but I sneak up on people without even trying.
>
> >Well, I was thinking more in terms of flexibility -- being able to fold
> >yourself in half to fit into the chest is a very nice way to hide.
>
> I think I've seen a system that put Stealth under Concentration
> skills. The ability to be careful seems more important than either
> agility or flexibility, although good balance is very helpful when
> freezing midstride. The situation you describe above would seem to me
> to involve use of a separate Contortionist skill rather than being
> straight hiding. Success at that skill then gives a large bonus to
> Hiding. Unless somebody tries to get something out of the chest.
>
> Anyway the real flexibility comes when you try to get out again! :-)

I should think that perception controls strealth more than
concentration or agility. Sure, speed can help move from one pillar
to the next, though this is more a combination of run & move silently,
and concentration would come into it more when multiple actions are
attempted (such as running/hiding or picking-lock/hiding. But to be
stealthy you need to pay attention to who is around you, what noises
you may make by performing certain actions, etc.

--
"The robot is going to lose. Not by much. But when the final score is
tallied, flesh and blood is going to beat the damn monster."
-Adam Smith

R Dan Henry

unread,
Aug 30, 2001, 1:44:39 AM8/30/01
to
On 28 Aug 2001 22:02:16 -0700, the disembodied brain of
mon...@robot.com (Robot Monster) transmitted thus:

>R Dan Henry <rdan...@earthlink.net> communicated:

>You could use several knowledge-type skills for this, such as:


>
>Knowledge: ___ Anatomy: ___ Xoznark
> | |_ Urrlak
> | \_ Zelion
> |
> |_ Culture ___ Xoznark
> | |_ Urrlak
> | \_ Zelion
> |
> \_ Communication ___ Xoznark
> |_ Urrlak
> \_ Zelion
>
>and then use these skills as modifiers when dealing with creatures of
>other species (in the same way as the proposed environment-advantage
>skills).

That would certainly be one way of doing it. I'm pretty set on just
branching out the individual skills by culture/race, tho.

>> Oh, well, could just do skill templates for
>> each race and profession and have all rookie Xoznark guard have the
>> same skills across the board. While unrealistic, you probably wouldn't
>> notice they all have the same Gambling with Yongrag Sticks skill
>> anyway.
>
>Is Yongrag Stick a race or a form of gambling? Or both? I'm
>picturing a version of cock-fighting using large stick insects.

I was imagining a sort of throwing-stuff game such as developed in
cultures the world over using marked sticks, carved bones, etc. Games
that appear to have grown out of divination and led ultimately to the
modern gamer's beautiful polyhedral dice sets. Although insect races
would be fun, too.

>> I'd thought of something like:
>> Biology----(given world)----(species of world). For a fantasy
>> world, I'd use the middle level for
>> mammal/reptile/avian/plant/fungus/etc.

>I'd not go into much detail further down the tree. It would take an
>awful lot of storage to retain data relating to anatomy knowledge for
>each individual animal and you wouldn't need more than three simple
>levels, eg. anatomy->reptile->lizard; anatomy->reptile->snake.

Pretty much. Ecosystems will be pretty simple, at least to begin with.
A few herbivores of various sizes, fewer carnivores, possibly an
intelligent species and associated domestic species. Similarly, only a
few plants per planet. The 'species' here really represent an entire
genus or even order in 'real-world' biology.

>Unless you intend on making a
>doctor/veterinarian class who can specialise in healing particular
>animals, this level of detail can be kept simple.

Well, medical science will go down to 'species' as well. But if the
setting were Earth, Zoology--Medicine--Earth--Dog would cover jackal,
foxes, domestic dogs, genetically enhanced talking dogs, and maybe
even hyenas.

>If you go down then to define skills such as Newt anatomy and Iguana
>anatomy you might upset players who love detail if you don't define
>the skills ToolManipulation->Fine Tools->Picking Locks->FiveTumblerKey
>and ToolManipulation->Fine Tools->Picking
>Locks->ThreeTumblerCombination : it is quite realistic to have an
>expert at one of these skills who is hopeless at the other. (But of
>course combination locks would be under perception anyway...)

LOL! "Sorry, mate, this here's a *three* tumbler lock and I'm a *five*
tumbler lock specialist. Guy you need is Three-Eyed Pete; you can find
him at Generic Spacedock Saloon #6. I'm outta here."

No, I don't want to split hairs that finely.

Of course, this is partly because most of the important stuff in
Nethack is either/or, whereas in Angband there are a good many numbers
that really count -- damage factors, speed, AC, stat bonuses, blows
per round, stealth, etc. In Nethack, you either are stealth or not.
Two levels of speed only. Damage hidden and not nearly so critical --
you can do stuff like swing a cockatrice corpse around and forget
damage dice. And in any case, you have normal weapons, and artifacts
and you'll see few artifacts, so simpler choices. You don't weigh a
Sword of Chaos (Holy Avenger) against your current artifact weapon
selection as in Angband. Angband is more quantitative, Nethack more
qualitative in the properties it gives characters and items.

Both approaches are valid, but lend themselves to different degrees of
player knowledge. Angband *requires* more player knowledge, Nethack
tends to suffer once too much knowledge is gained because so much of
what's cool about it is finding stuff out.

>> What shouldn't happen is misleading things. If your bonuses work in a
>> non-linear fashion, so that a Knife (+10) may not give you twice the
>> bonus of a Knife (+5), don't display numbers. Call them Steel Knife
>> and Iron Knife instead and just offer a list showing the order of the
>> value of metals (or make that available within the game).
>
>Of course using alloys as weapon prefixes would mean renaming scrolls
>of enchant weapon to scrolls of upgrade material. Unless you pulled
>the console RPG trick and simply set weapon quality as fixed (ie. the
>only way to get a better weapon than a ceramic sword is to ditch it
>and buy a titanium sword or (better yet!) a Laconian sword.

Well if it's a matter of reading scrolls over it, have cutoff points
where you show improvements like:

Unenchanted -> Mildly Enchanted -> Enchanted -> Richly Enchanted ->
Deeply Magical

The point is not to offer numbers that appear to have a meaning which
they do not have.

>> >I'm still not quite clear how you will model physical attributes in a
>> >skill-only system -- is it all hidden directly in the skills, or is there
>> >some upper-level "root" for the skills (and/or display)?
>>
>> If your character can do a lot of strong-guy stuff, you can call him
>> strong. There will be some relationships between many of these skills
>> as they'll be in the same skill trees. Attributes like height, weight,
>> and eye color are handled normally. But the game won't give you a
>> value for traditional stats, just skill levels. Some of these are
>> going to be fairly general. If you look at 'lifting', 'carrying', and
>> 'striking' skills, you can justly call someone with high scores in all
>> of these strong, as someone who in the real world did lots of lifting,
>> carrying, and striking would be strong. But a low skill might just
>> indicate a poor application of existing muscle strength, so it isn't
>> just a split-up stat.
>
>If there is any way in the game of increasing/decreasing muscle
>strength other than through exercise (ie. through magic) then you
>should have some way of mapping this to your skill set.

Well, there really won't be in the system I'm planning for, as the
setting is sci-fi and extraordinary boosts would be things like
cybernetics or 'rapid learning' systems that would be focused on
enhancing skills to begin with.

Which isn't to say I couldn't adapt it to a magical world, it's just
that the results might not be what you're accustomed to. I don't think
that's such a bad thing.

> You note that
>skills like lifting are not just a split-up of ability and technique.
>Could two characters exist who could lift the same weight where char1
>did it using brute strength and char2 did it using a combination of
>strength and lifting techniques? If so, would a potion of strength
>give more of a boost to the first character, because the second relies
>less on strength? Would a potion of Forget Climbing Ability affect
>them equally?

Magic uses existing game mechanics, so the same plus or minus applies
in all cases (although racial bonuses won't be reduced if minimal
skill is 0 -- so a centaur could drink many Potions of Poor Carrying
and still haul around a decent amount of stuff). The closest
equivalent of 'strength' would be the combination of various basic
physical feat skills that are more 'brawny' than 'agile'. So you gulp
down Zaroonderel's Elixir of Brawniness and get a bonus to 'lifting',
'striking', 'carrying', and maybe 'bending' if you want a skill like
that. If you have several 'strength-themed' items, you might want to
create a template for 'Strength' as an aptitude. Personally, I'd have
a Girdle of Mesa Giant Strength improve Striking and Lifting skills,
but *not* Carrying, as giants don't really seem to be great carriers
-- maybe a sack now and then, but using their bulk more than their
might for such tasks. Whereas giants are well known for picking up
heavy things, throwing them (okay, need to have the Girdle add to
Throwing as well) and smashing things. But generally, once I get rid
of stats, I'm not terribly interested in sneaking them back in, but
they can be modeled and quite flexibly.

I never really thought of strength-enhancing magic as building muscle
mass anyway, but rather as enchanting the individual, adding a sort of
'aura of might'. So it can work whatever way is convenient as far as
I'm concerned.

>> Off the top of my head, without having actually worked out a set of
>> skill trees, I'd say probably something like this:
>>
>> Tool Manipulation----Large Tools----Smithing----Etc.
>> \_Fine Tools----Engraving
>> \_Jeweler
>> |_Picking Locks
>> |_Making Locks
>>
>
>The problem with skill trees rears its ugly head. Jeweler might be
>better off being a Art/Craft skill with a bonus given from Fine Tools
>skill.

What art/craft skills are you thinking of that wouldn't be in the tool
manipulation tree? Paintbrushes are fine tools, looms large ones.
Basketry might be problematic... Again, I'm giving only provisional
answers, I couldn't say just how the skill trees would look until I
actually made the game.

For example, I need to know enough about the mechanics I'll use to
know if different skill trees can be of arbitrary depth or if all
trees need to be made to be four layers deep for the numbers to work
out well.

>> I think I've seen a system that put Stealth under Concentration
>> skills. The ability to be careful seems more important than either
>> agility or flexibility, although good balance is very helpful when
>> freezing midstride. The situation you describe above would seem to me
>> to involve use of a separate Contortionist skill rather than being
>> straight hiding. Success at that skill then gives a large bonus to
>> Hiding. Unless somebody tries to get something out of the chest.
>>
>> Anyway the real flexibility comes when you try to get out again! :-)

>I should think that perception controls strealth more than
>concentration or agility. Sure, speed can help move from one pillar
>to the next, though this is more a combination of run & move silently,
>and concentration would come into it more when multiple actions are
>attempted (such as running/hiding or picking-lock/hiding. But to be
>stealthy you need to pay attention to who is around you, what noises
>you may make by performing certain actions, etc.

It could be argued that perception should fall into concentration
skills, as well. In any case, I'm looking at what causes failures and
usually it's a lapse of concentration -- although that often leads to
a failure of perception. Actually, yes, I think perception skills
should go under concentration as they require the ability to focus on
one's surrounding without being distracted.

Archibald

unread,
Aug 30, 2001, 4:38:29 PM8/30/01
to
[...]

>and IMO having two skills that do one thing is something worth avoiding.
[...]

>I also agree that duplicating skills is *usually* to be avoided.
[...]

I cannot agree. Did You seen "Crawl"? There are only three kind of skills
(maybe few others too).

1) bonus to HIT or DAM - Fighting, Short Sword, Long Swords, Axes....
2) bonus to AC or EV - Shield, Armour, Dodge
3) magic bonus - Spellcraft, Enchantment, Air/Fire/Cold...

All this skills do this same, but with slighty diffirence. You can train
'Fighting' to gain small bonus for everything or 'Axes' to gain this same
bonus but greater and only for axes.

This works very well.


Archibald

Robot Monster

unread,
Aug 31, 2001, 12:30:04 AM8/31/01
to
R Dan Henry <rdan...@earthlink.net> communicated:
>
> On 28 Aug 2001 22:02:16 -0700, the disembodied brain of
> mon...@robot.com (Robot Monster) transmitted thus:
>
> >If you go down then to define skills such as Newt anatomy and Iguana
> >anatomy you might upset players who love detail if you don't define
> >the skills ToolManipulation->Fine Tools->Picking Locks->FiveTumblerKey
> >and ToolManipulation->Fine Tools->Picking
> >Locks->ThreeTumblerCombination : it is quite realistic to have an
> >expert at one of these skills who is hopeless at the other. (But of
> >course combination locks would be under perception anyway...)
>
> LOL! "Sorry, mate, this here's a *three* tumbler lock and I'm a *five*
> tumbler lock specialist. Guy you need is Three-Eyed Pete; you can find
> him at Generic Spacedock Saloon #6. I'm outta here."
>

Put that into a game and I'd buy it!

> >>
> >> Tool Manipulation----Large Tools----Smithing----Etc.
> >> \_Fine Tools----Engraving
> >> \_Jeweler
> >> |_Picking Locks
> >> |_Making Locks
> >>
> >
> >The problem with skill trees rears its ugly head. Jeweler might be
> >better off being a Art/Craft skill with a bonus given from Fine Tools
> >skill.
>
> What art/craft skills are you thinking of that wouldn't be in the tool
> manipulation tree? Paintbrushes are fine tools, looms large ones.
> Basketry might be problematic... Again, I'm giving only provisional
> answers, I couldn't say just how the skill trees would look until I
> actually made the game.


That's like saying that coding should be in the keyboard manipulation
tree. Yes, tool manipulation is involved, but the thought processes
involved in lock picking and creating jewellery are entirely
different. Crafting something from relies a lot on design ability,
whether you are fashioning a ring or a song or a piece of software.
If you were simply copying something else, *then* you would only need
to be good at the tool manipulation.

Only when you aren't any good. When you have a low skill score (at
anything mind you, not just perception) you need to concentrate more
on the activity. As you become more proficient you need less
concentration to do a good job. I'm sure you agree that routine tasks
require little attention be paid to be successful, whereas if you are
doing something that challenges you you generally only succeed when
you give it your full mental attention.

R Dan Henry

unread,
Aug 31, 2001, 4:25:59 AM8/31/01
to
On Thu, 30 Aug 2001 22:38:29 +0200, the disembodied brain of Archibald
<kko...@kki.net.pl> transmitted thus:

>[...]
>>and IMO having two skills that do one thing is something worth avoiding.
>[...]
>>I also agree that duplicating skills is *usually* to be avoided.
>[...]
>
>I cannot agree. Did You seen "Crawl"? There are only three kind of skills
>(maybe few others too).

Neither of those statements above is a universal claim, merely
general, and cannot be disproven by a single counter-example.
Furthermore, in the sentence you failed to properly attribute to me, I
stressed that this was *usually* the case, indicating that I do
believe there are cases where duplication is justified.

>1) bonus to HIT or DAM - Fighting, Short Sword, Long Swords, Axes....

Short Sword skill and Axe skill do not duplicate each other. Axes and
short swords are different weapons.

>2) bonus to AC or EV - Shield, Armour, Dodge

Skill in dodging and skill using armor do not duplicate each other.
They are different forms of defense.

>3) magic bonus - Spellcraft, Enchantment, Air/Fire/Cold...

Skill with spell type A does not duplicate skill with spell type B.
The whole point of having spell categories is that they use different
skills.

>All this skills do this same, but with slighty diffirence. You can train
>'Fighting' to gain small bonus for everything or 'Axes' to gain this same
>bonus but greater and only for axes.

Similar purposes is not at all the same as skill duplication. And
'Fighting' as you describe it here is a 'higher level' skill on the
skill tree, something that has been assumed will exist throughout this
entire conversation.

>This works very well.

I reserve judgment, as I haven't played much Crawl. However, you are
mistaking similarity of purpose for skill duplication and they are not
the same thing at all. Skill duplication is when two skills do *the
same thing* or have a large overlap. This does not count the
relationships between levels of skill trees, which are designed to
work together systematically. What you have described here is simply
having different skills. "Sword" skill may *increase combat
effectiveness* with swords, while "Axe" skill may *increase combat
effectiveness* with axes, but this is not skill duplication, the *with
swords* and *with axes* parts are as important to defining the skills
as the *increase combat effectiveness* part.

R Dan Henry

unread,
Aug 31, 2001, 4:26:05 AM8/31/01
to
On 30 Aug 2001 21:30:04 -0700, the disembodied brain of

mon...@robot.com (Robot Monster) transmitted thus:

>R Dan Henry <rdan...@earthlink.net> communicated:
>>
>> On 28 Aug 2001 22:02:16 -0700, the disembodied brain of
>> mon...@robot.com (Robot Monster) transmitted thus:

>> >>

>> >> Tool Manipulation----Large Tools----Smithing----Etc.
>> >> \_Fine Tools----Engraving
>> >> \_Jeweler
>> >> |_Picking Locks
>> >> |_Making Locks
>> >>
>> >
>> >The problem with skill trees rears its ugly head. Jeweler might be
>> >better off being a Art/Craft skill with a bonus given from Fine Tools
>> >skill.
>>
>> What art/craft skills are you thinking of that wouldn't be in the tool
>> manipulation tree? Paintbrushes are fine tools, looms large ones.
>> Basketry might be problematic... Again, I'm giving only provisional
>> answers, I couldn't say just how the skill trees would look until I
>> actually made the game.

>That's like saying that coding should be in the keyboard manipulation
>tree. Yes, tool manipulation is involved, but the thought processes
>involved in lock picking and creating jewellery are entirely
>different. Crafting something from relies a lot on design ability,
>whether you are fashioning a ring or a song or a piece of software.
>If you were simply copying something else, *then* you would only need
>to be good at the tool manipulation.

Design would be a separate skill. And subjective, so I'd probably put
it under social skills, as the ability to create a design pleasing to
Sewer Trolls does not imply the ability to create a design pleasing to
Crystal Gnomes. Or possibly there should be an "Art" tree branching
out by medium and culture. In any case, actually *realizing* the
design would be tool manipulation skill.

Note that most of the time, I wouldn't expect a character in a
roguelike to be too concerned with how artsy his work was. Jeweler
skill would be typically used for working magical rings and amulets,
etc. The engraving skill above would probably be used for
forgery/counterfeiting, which isn't too original.

Also bear in mind that most cultures did not place the same value on
'originality' that modern industrialized society does. In most
situations, a well-executed version of a traditional pattern is just
fine.

>> >I should think that perception controls strealth more than
>> >concentration or agility. Sure, speed can help move from one pillar
>> >to the next, though this is more a combination of run & move silently,
>> >and concentration would come into it more when multiple actions are
>> >attempted (such as running/hiding or picking-lock/hiding. But to be
>> >stealthy you need to pay attention to who is around you, what noises
>> >you may make by performing certain actions, etc.
>>
>> It could be argued that perception should fall into concentration
>> skills, as well. In any case, I'm looking at what causes failures and
>> usually it's a lapse of concentration -- although that often leads to
>> a failure of perception. Actually, yes, I think perception skills
>> should go under concentration as they require the ability to focus on
>> one's surrounding without being distracted.
>
>Only when you aren't any good. When you have a low skill score (at
>anything mind you, not just perception) you need to concentrate more
>on the activity. As you become more proficient you need less
>concentration to do a good job. I'm sure you agree that routine tasks
>require little attention be paid to be successful, whereas if you are
>doing something that challenges you you generally only succeed when
>you give it your full mental attention.

I don't fully agree. Yes, certain skills allow for routine work
without concentration. Observation skills are not among them. Insofar
a stealth is observation-related, it follows that it, too, requires
concentration even when skilled. Thinking as you do causing traffic
accidents when drivers decide they don't need to be concentrating
because they have developed proficiency. Unfortunately, observation is
a fair chunk of driving. Not that you don't get help from learning
what to look for, but lapses of concentration continue to be the point
of failure. More so for experienced individuals than for novices, who
will make more mistake through not knowing just what to do in the
first place.