Scoring, and Time limits

[Guest]

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

For those of you lucky enough to have tried Steamband, you may have
noted I'm really trying to take it to the next level. (That's hyperbole)
There are enough carbon copy variants of angband, and I'd like to make
one that is a different play experience. Currently there is something
called 'Steamware' which allows you to research and then install
components into your 'body' giving you innate powers. The research takes
money and a set amount of time, after which you can get it installed.

Here is the deal. The 'set amount of time' is meaningless. Anyone can
just buy it and then rest until it is ready. I see two solutions to the
problem.

1) Create a scoring system that rewards good play, and penalizes
excessive turn count.

and

2) Create a time limit.

What I'd really like from the community are some suggestions on how to
do the above. And they are closely releated - you can give points for
killing monsters, but only if you don't have an unlimited amount of time
to do so. What are good ways to do these things.

Comments and disscussion exteremely welcome.
-Campbell

 

[Guest]

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

Courtney Campbell <campbell@oook.cz> writes:

> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.
>
> and
>
> 2) Create a time limit.
>
> What I'd really like from the community are some suggestions on how to
> do the above. And they are closely releated - you can give points for
> killing monsters, but only if you don't have an unlimited amount of time
> to do so. What are good ways to do these things.

I am all in favor of encouraging people to play more quickly, but I
don't like the idea of forcing someone else to use an artificial time
limit. Different race/class combinations need really different
limits. As to scoring systems, I know I just ignore the score in
so I wouldn't be surprised if others simply ignore your new scores.

There is a third alternative. Offer fixed resources for the duration
of an entire game. For example, offer 99 !CCW in the temple, and
that's it. People who take too long run out. I was loudly shouted
down the last time I suggested this approach, but I still like it.


Eddie

 

[Guest]

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

Courtney Campbell wrote:
> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.
>
> and
>
> 2) Create a time limit.
>
> What I'd really like from the community are some suggestions on how to
> do the above. And they are closely releated - you can give points for
> killing monsters, but only if you don't have an unlimited amount of time
> to do so. What are good ways to do these things.
>
> Comments and disscussion exteremely welcome.
> -Campbell

You could base the research time off of something else instead.
If you based it off of depth then it would actually be a finite
resource, and the player would have to make harder choices about what
to research. (maybe the research doesn't complete until the player
visits a depth that is current depth +2)
You could also base it off of the player's exp, so they have to go
fight in order for the research to proceed. That might penalize the
heavy divers that spend more time grabbing treasure than clearing pits
though.

Scott

 

[Guest]

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

On 2005-07-12, Courtney Campbell <campbell@oook.cz> wrote:
> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.
>
I prefer this one, it would reward fast players and don't drive away
slow players.

> and
>
> 2) Create a time limit.
>
I would probably stop playing a variant with a time limit, my playing
style is quite slow and I often clear a level about five times before
I go to the next one.

> What I'd really like from the community are some suggestions on how to
> do the above. And they are closely releated - you can give points for
> killing monsters, but only if you don't have an unlimited amount of time
> to do so. What are good ways to do these things.
>
If you really want a time limit, the only type that make sense to me is
aging. This would probably involve stats that starts to drop as the character
age and if a stat goes to zero it would be game over. (The player goes
senile if it was a mental stat and the body gives up if it was a physcial
stat.)


> Comments and disscussion exteremely welcome.
> -Campbell

--
Christer Nyfalt

 

[Guest]

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

In article <m3y88ct6mw.fsf@hot.NOSPAM.mail.com>,
Eddie Grove <eddiegrove@hot.NOSPAM.mail.com> wrote:

> I am all in favor of encouraging people to play more quickly, but I
> don't like the idea of forcing someone else to use an artificial time
> limit. Different race/class combinations need really different
> limits. As to scoring systems, I know I just ignore the score in
> so I wouldn't be surprised if others simply ignore your new scores.

This, for example, in crawl is handled by having different races and
classes have different digestion rates. Whatever the 'time limit' is, it
should be tuned to the various races and classes (Steam-Mecha should not
have a very long time limit for instance)

I believe people ignore the score because it doesn't provide any
meaningful info. If it literally represented playing skill, people might
use it to, say, compare their characters on a high score list.

What exactly is on the list of things that defines skillful play?

> There is a third alternative. Offer fixed resources for the duration
> of an entire game. For example, offer 99 !CCW in the temple, and
> that's it. People who take too long run out. I was loudly shouted
> down the last time I suggested this approach, but I still like it.

Sort of a reverse time limit? I sort of view it as six of one, half a
dozen of the other. If the 'time limit' restricts you to 123 !CCW, or
the store eventually runs out of them after 123 are sold, what is the
quantifiable difference?

I don't think anyone should be shouted at for suggesting radical
changes. [V] is always going to be [v], radical changes to the game
aren't going to ruin [v], and there's no reason a popular variant has to
play exactly like [V].

Hmmm. I think I like that statment. Why, I like all kinds of games,
Wizadry, Final Fantasy, Disgaea - why is it a requirement that someone
working on a variant produce something that's like [V]? Especially when
[v] exists?

THERE IS NO REASON A VARIANT HAS TO PLAY, LOOK, OR ACT ANYTHING LIKE [V].

Ah, yes, I feel much better now. I'd love some radical ideas on what
'good play' is (so we can score it), and radical ideas on a time limit.

> Eddie

Thanks Eddie. The rant above is *obviously* not directed at you.
-Campbell

 

[Guest]

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

In article <1121181622.095055.326750@g14g2000cwa.googlegroups.com>,
"Scott Yost" <yostage@gmail.com> wrote:

> You could base the research time off of something else instead.
> If you based it off of depth then it would actually be a finite
> resource, and the player would have to make harder choices about what
> to research. (maybe the research doesn't complete until the player
> visits a depth that is current depth +2)

I like depth as a finite resource, espically in Steamband where there
are only 50 dungeon levels, but I don't think Steamware is the place for
it.

What are some other places depth could be used as a resource?
-Campbell

> You could also base it off of the player's exp, so they have to go
> fight in order for the research to proceed. That might penalize the
> heavy divers that spend more time grabbing treasure than clearing pits
> though.
>
> Scott

 

[bill]

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

Courtney Campbell wrote:
> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.
>
> and
>
> 2) Create a time limit.
>
> What I'd really like from the community are some suggestions on how to
> do the above. And they are closely releated - you can give points for
> killing monsters, but only if you don't have an unlimited amount of time
> to do so. What are good ways to do these things.
>
> Comments and disscussion exteremely welcome.
> -Campbell

Fallout had a quest with a time limit, you could still win the game but
you missed seeing one of the endings if you took too long. It was
immensly unpopular; for Fallout II the developers went to great pains
to assure everyone that they could take as long as they wanted and
still experience all of the gameplay.

In contrast, Wizardry 6? - Crusaders of the Dark Savant had some quests
where you were competing with NPCs to find objects needed to complete
the game. If you took too much time and they found the objects first,
you had to find them and obtain the objects.

I mention these because they were both very popular games, suggesting
that you don't have a hard time limit but instead make Steamware a bit
harder or more involved to obtain if the player takes too much time.

As you say, using game turns as a measure of research time doesn't
reward players for boldness. One thought might be to use dungeon level
instead, buy the tech at one level and have it activate 5 or 10 levels
deeper. If that's not enough incentive you might put a time limit on
the activation; if the player doesn't dive fast enough the tech
vanishes and has to be bought again.

Building a scoring system is the more difficult of your two tasks. The
current measure is pretty broken when the leader on the Angband ladder
got there by playing for 1 1/2 years so he could accumulate 99,999,999
EXP. From what I've read on this newsgroup, the things that people are
really impressed with are:

Did you defeat Morgoth?
Did you do it in a very few turns, or at CL < 50? (Eddie)
Did you handicap yourself (artifactless, ironman)? (Timo)

The trick would to quantify the above. And you won't know if your sceme
is really balanced until enough players have finished the game and
given you some scores. If I ever get around to making a variant, here's
a possibility I might try:

Start the player with an arbitrarily high score, then subtract game
turns from the score. Make the starting score higher for classes that
just can't dive as fast. Also subtract some multiple of CL to encourage
finishing before CL50. Make the starting value higher if artifactless,
etc. is selected. Then add something at the end when the big boss is
defeated.

Bill

 

[Guest]

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

In article <1121205702.107514.12380@g44g2000cwa.googlegroups.com>,
"Bill" <wpeterson1@socal.rr.com> wrote:

> Building a scoring system is the more difficult of your two tasks. The
> current measure is pretty broken when the leader on the Angband ladder
> got there by playing for 1 1/2 years so he could accumulate 99,999,999
> EXP.

Right!

> From what I've read on this newsgroup, the things that people are
> really impressed with are:
>
> Did you defeat Morgoth?
> Did you do it in a very few turns, or at CL < 50? (Eddie)
> Did you handicap yourself (artifactless, ironman)? (Timo)

There apparently is a sweet spot for turns. We also have to look at
things like connected stairs (should it even be an option). And there is
sort of a 'sweet spot' in turns for each playing style. The handicaps
should definately be some kind of bonus to score.

> The trick would to quantify the above.

Well, Yes. Lets. Lots of people here, knowing lots about this (and this
type) of game.

1) Unique(s) defeated.
2) Turn count (absolute or player? I would say it would almost have to
be player being how much absolute depends on speed items)
3) Character Level (lower is better, natch)
4) Handicaps
o) Artifactless
o) Ironman
o) ?


> And you won't know if your sceme
> is really balanced until enough players have finished the game and
> given you some scores.

Well, that's certainly not a reason to not do something. I'll be
glad to tweak and balance the system till it works right.

> If I ever get around to making a variant, here's
> a possibility I might try:
>
> Start the player with an arbitrarily high score, then subtract game
> turns from the score. Make the starting score higher for classes that
> just can't dive as fast. Also subtract some multiple of CL to encourage
> finishing before CL50. Make the starting value higher if artifactless,
> etc. is selected. Then add something at the end when the big boss is
> defeated.

Score as total that decreases, with bonuses along the way.

vs.

Score that increases.

Perhaps there could be two halves - one score that increments as you
play, and a 'death score' that takes your final turn count + character
level into account.

> Bill

-Campbell

 

[Guest]

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

Courtney Campbell wrote:
> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.
>
> and
>
> 2) Create a time limit.
>

3) Why research 1st and install then? Buy and install the Steamware.
After it is installed, you learn to use it: gain set amount of experience
before it works fine. While learning to use steamware, experience goes
to steamware use, not player levels.

Experience is meaningless when you are a few million points past level 50,
so this is a halfway solution.

A scoring system would be interesting, but would it be used?
I think many people play to win, not to get a score.

 

[Guest]

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

In article <db0d37$v6$1@oravannahka.helsinki.fi>,
Juho Schultz <juho.schultz@helsinki.fi> wrote:

> Courtney Campbell wrote:
> > For those of you lucky enough to have tried Steamband, you may have
> > noted I'm really trying to take it to the next level. (That's hyperbole)
> > There are enough carbon copy variants of angband, and I'd like to make
> > one that is a different play experience. Currently there is something
> > called 'Steamware' which allows you to research and then install
> > components into your 'body' giving you innate powers. The research takes
> > money and a set amount of time, after which you can get it installed.
> >
> > Here is the deal. The 'set amount of time' is meaningless. Anyone can
> > just buy it and then rest until it is ready. I see two solutions to the
> > problem.
> >
> > 1) Create a scoring system that rewards good play, and penalizes
> > excessive turn count.
> >
> > and
> >
> > 2) Create a time limit.
> >
>
> 3) Why research 1st and install then? Buy and install the Steamware.
> After it is installed, you learn to use it: gain set amount of experience
> before it works fine. While learning to use steamware, experience goes
> to steamware use, not player levels.

Thanks for the suggestion, but without question, steamware is stay as it
is. It currently works like research in an rts, and is fullfilling it's
role quite nicely.

I see by providing background information, I've made it well-neigh
impossible to get feedback on my original questions. I'm interested in
suggestions for a scoring system, and ideas for a (soft/hard) time
limit. (such as crawl using hunger, and adom using mutations).

Personally, I think the best time limit available is one that wants to
make you finish quicker, instead of one that penalizes you for taking
too long. At some point, you baseline because you were too slow (hence
how scoring came up as one of the available options)

> Experience is meaningless when you are a few million points past level 50,
> so this is a halfway solution.

Yes, well, that's why we're having this disscussion. Score, experience,
turn count, monsters killed, all these things (and more!) are
meaningless, because they simply represent tedium instead of skill.

> A scoring system would be interesting, but would it be used?
> I think many people play to win, not to get a score.

I don't really care about wether it would be 'used' or if people would
play to win, or get a score. If it were an accurate representation of
player skill, and progress in the game, it would be used, because it
would be *useful*.
-Campbell

 

[Guest]

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

Courtney Campbell writes:
> For those of you lucky enough to have tried Steamband, you may have
> noted I'm really trying to take it to the next level. (That's hyperbole)
> There are enough carbon copy variants of angband, and I'd like to make
> one that is a different play experience. Currently there is something
> called 'Steamware' which allows you to research and then install
> components into your 'body' giving you innate powers. The research takes
> money and a set amount of time, after which you can get it installed.
>
> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.
>
> 2) Create a time limit.
>
> What I'd really like from the community are some suggestions on how to
> do the above. And they are closely releated - you can give points for
> killing monsters, but only if you don't have an unlimited amount of time
> to do so. What are good ways to do these things.

I don't think setting a hard time limit is good. Ragnarok sort of
pulled it off, but even there, there was an item that could help out a
lot. But soft time-incentives are great. In my dream roguelike, which
I will write some day in the distant future, there will be several.

First, you have the bonus stage incentive. Put something after the end
of the game, sort of like the lab. Make it worth playing, something a
bit different, but not necessary to feel you've won. And you have to
get there before a certain time limit or you don't get in.

Second, positive reinforcement. Put in a kind of item or other bonus
that doesn't show up in the very early game, but is totally unavailable
past a time limit. E.G. There are recent visitors from the surface
stranded in the higher levels with technology to improve your stuff.
But if you don't rescue them in time, they'll all be killed off.

Third, negative reinforcement. The martians don't land until a certain
time has passed, but once they do, you can encouter martians, and
they're a bit tougher than most. Or some evil cult succeeds in their
ritual, causing the whole world to be filled with slightly dangerous
heat, or disease, or maybe destroying the ability for anyone to
teleport.

Fourth, the soft deadline. You're going to die or irrevocably fail if
you take too long. But you can push it back. Maybe you're dying of
sunlight-deprivation, but if you can find enough sunlight generator
mechanisms, you'll pull through. Or maybe the bottom of a volcano is
going to leak through flooding the center of the earth unless you
periodically release the pressure, but that requires going somewhere
dangerous. This one is probably the trickiest to do. You don't want
it to require too much sheer luck to extend the game, but then, you
don't want it to just be a formality of picking up the needed item
either.


Then there is the matter of character speed. You don't want your
limits to be meaningless for fast characters, and horrible for slow
ones. I see a few solutions to this problem.

Don't make character speed vary that much. Obviously, this isn't the
case in Steamband as is, and you probably don't want to change the game
that much. But it is possible for say, 2x speed to be about as fast as
you can go and still have speed be very valuable in a game. See Crawl.

Make speed cost something meaningful. Maybe going fast for long
periods of time mutates you. Or calls down divine wrath. Or gives
power to some late game unique who draws power from the speed of
others.

Ignore speed. If the time limit is a disease that the player has,
maybe speed just speeds up the disease to match. If it's based on
events outside the player, maybe you just arbitrarily make them match
up because the character doesn't really know when they're going to
happen anyway.


Personally, I'd love to see a stage-based reward and penalty based time
system in steam. First you have to kill some unique in a certain
amount of time to get a reward (perhaps an extra skill or steamware is
available). Once you've killed him, whether or not it was in time, you
have to kill a bunch of some type of enemy to avoid changing the laws
of physics against you mildly for the rest of the game. Etc. Plus a
bonus stage at the end based on time. Maybe if you make it in time you
go to the lab, but if you don't, you go to some horribly scarred earth.
And either make it based on player turns, or make each stage scale to
the highest speed the player had so far in the game. So if you find a
speed item, you can use it, or you can save it to make the next stage
easier.

 

[Guest]

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

"Courtney Campbell" <campbell@oook.cz> schrieb...

> Here is the deal. The 'set amount of time' is meaningless. Anyone can
> just buy it and then rest until it is ready. I see two solutions to the
> problem.

As Juho Schultz wrote, you could make it dependant on experience instead
of time (was my first thought, too).

Or the modules could start out at a low power rate or high fail rate,
and become more better when using the module often (but players might
find cheap tricks to train the module in save situations).

> 1) Create a scoring system that rewards good play, and penalizes
> excessive turn count.

Low turn count doesn't necessary mean good play.
With connected stairs, you can < > to gain many items and experience in
a short time, but is this good play?
(i sometimes do this with low level mages, but they have a high turn
count since the have to rest very often).

> 2) Create a time limit.

So it's best to quit your character if you don't get a speed item
before 2000'.

Werner.

 

[Guest]

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

In article <3ji6leFq95a8U1@individual.net>,
Werner Bär <werner.baer@gmx.de> wrote:

> > 2) Create a time limit.
>
> So it's best to quit your character if you don't get a speed item
> before 2000'.

This is definately part of the issue. We could look at it like this
however, absolute time is *player* turns, all speed does is increase
your reflexes making monsters sluggish, giving you more opportunities
for actions.

Why would scoring, or the time limit be based off anything *but* player
turns? Reality? In fact, this very second, I think I'm going to alter
steamware so that it's based off player turns.

> Werner.
-Campbell

 

[Guest]

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

Courtney Campbell <campbell@oook.cz> writes:

> In article <3ji6leFq95a8U1@individual.net>,
> Werner Bär <werner.baer@gmx.de> wrote:
>
>> > 2) Create a time limit.
>>
>> So it's best to quit your character if you don't get a speed item
>> before 2000'.
>
> This is definately part of the issue. We could look at it like this
> however, absolute time is *player* turns, all speed does is increase
> your reflexes making monsters sluggish, giving you more opportunities
> for actions.
>
> Why would scoring, or the time limit be based off anything *but* player
> turns? Reality? In fact, this very second, I think I'm going to alter
> steamware so that it's based off player turns.

I've lost track of what was intended, and I have never played Steam,
but nevertheless I have one more suggestion.

There are times when you don't want the player to be able to "scum" in
whatever sense by resting on dLevel 1. I suggest that in those
instances instead of counting turns, you count number of squares
explored, perhaps multiplied by a difficulty factor [exponential in
the level?].


Eddie

 

[Guest]

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

In article <campbell-2A2DC4.03495012072005@news.isp.giganews.com>,
Courtney Campbell <campbell@oook.cz> wrote:
>For those of you lucky enough to have tried Steamband, you may have
>noted I'm really trying to take it to the next level. (That's hyperbole)
>There are enough carbon copy variants of angband, and I'd like to make
>one that is a different play experience. Currently there is something
>called 'Steamware' which allows you to research and then install
>components into your 'body' giving you innate powers. The research takes
>money and a set amount of time, after which you can get it installed.
>
>Here is the deal. The 'set amount of time' is meaningless. Anyone can
>just buy it and then rest until it is ready. I see two solutions to the
>problem.
>
>1) Create a scoring system that rewards good play, and penalizes
>excessive turn count.

For a start on this, it seems like you may as well eliminate the
middleman and have steamware reduce score directly. I'm all for
revamping the score system, but haven't got any ideas on that, either.
I like having some time-lag on the steamware, but it wouldn't need to be
as long as it is currently.

>2) Create a time limit.

My knee-jerk reaction to this is always "no!", but having played a
little more Crawl and read a bit more on r.g.r.dev, I realize that this
is just my Angband sensibilities talking: the Angband philosophy has
always seemed to be "as much time as you need" (and this is one of the
things that I enjoy about the game, though the thought of a time limit
scares me a lot less than it used to).

Despite a further lack of concrete ideas, I'll note that a hard turn
limit is probably a poor idea, since it will either be a non-factor or
be crippling, depending on one's rate of play (and other factors, as
mentioned by Werner and Eddie).

-Andrew ()

 

[Guest]

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

"Igor D. WonderLlama" <WonderLlama_@hotmail.com> wrote in
news:1121207619.111233.72480@g49g2000cwa.googlegroups.com:

> You don't want
> it to require too much sheer luck to extend the game, but then, you
> don't want it to just be a formality of picking up the needed item
> either.

Such as food and light generally are in Angband outside of Ironman
play. Sure, you can starve to death, but who really does outside of
constant paralysis?

 

[magnate]

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

Courtney Campbell wrote:
> "Bill" <wpeterson1@socal.rr.com> wrote:

> > Building a scoring system is the more difficult of your two tasks. The
> > current measure is pretty broken when the leader on the Angband ladder
> > got there by playing for 1 1/2 years so he could accumulate 99,999,999
> > EXP.

> Right!

Agreed.

> > From what I've read on this newsgroup, the things that people are
> > really impressed with are:
> >
> > Did you defeat Morgoth?
> > Did you do it in a very few turns, or at CL < 50? (Eddie)
> > Did you handicap yourself (artifactless, ironman)? (Timo)

This made me laugh. So simple, yet so true.

> > The trick would to quantify the above.

> Well, Yes. Lets. Lots of people here, knowing lots about this (and this
> type) of game.
>
> 1) Unique(s) defeated.

All uniques are different, and worth differing contributions to score.
The differences are quantified in their xp value. You can't simply use
xp for score, because of farming etc. But you could keep track of
"unique xp" separately, and that would be a useful element of a score.

> 2) Turn count (absolute or player? I would say it would almost have to
> be player being how much absolute depends on speed items)

Very definitely player turns. Finding speed items is a matter of luck,
not skill. IMHO the relationship between player turns and skill has a
big enough linear region to make its inclusion in a score quite easy.
Only when comparing Eddie with Lev would you need to fine tune the
exponent ...

> 3) Character Level (lower is better, natch)

This is really a function of turn count, and we should be wary of
allowing it too much value as an independent component of score. Also,
the cap makes it horrible to include in a smooth scoring algorithm. It
would be better to use total xp instead, as this keeps increasing after
clev 50. If you decide to include it as a separate element, simplest
would be to make it something like 50 million minus total xp. It's not
a linear progression though (and nor is clev, btw), so you'd want
something more like 10,000-sqrt(xp).

[Hmm, you'd then need to sqrt the "unique xp" mentioned above to
harmonise the components.]

> 4) Handicaps
> o) Artifactless
> o) Ironman
> o) ?

Egoless. Bookless. Missile-less. Naked. All sorts of restrictions are
possible. Each would need careful calibration to fit into the scoring
algorithm. Are they additive (a simple addition to the total score)? I
suspect they would work better as multiplicative, eg.

Ironman: 2x
Artifactless: 1.5x
Egoless: 1.8x
Bookless: 1.7x (mages), 1.4x (priests)
Missile-less: 1.2x
Naked: 10x (warrior), 8x (ranger, rogue, paladin), 3x (mage, priest)

.... you get the idea. These numbers are just off the top of my head and
would need much debate.

> Score as total that decreases, with bonuses along the way.
>
> vs.
>
> Score that increases.
>
> Perhaps there could be two halves - one score that increments as you
> play, and a 'death score' that takes your final turn count + character
> level into account.

This seem unnecessarily complex. I can't see any reason not to have a
"score" field which is updated every turn. Each turn you don't kill any
uniques it drops a tiny bit. Each time you kill a unique it rises a
bit. When you kill Morgy it rises quite a lot.

CC

 

[Guest]

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

Billy Bissette writes:
> "Igor D. WonderLlama" <WonderLlama_@hotmail.com> wrote in
> news:1121207619.111233.72480@g49g2000cwa.googlegroups.com:
>
> > You don't want
> > it to require too much sheer luck to extend the game, but then, you
> > don't want it to just be a formality of picking up the needed item
> > either.
>
> Such as food and light generally are in Angband outside of Ironman
> play. Sure, you can starve to death, but who really does outside of
> constant paralysis?

Exactly what I was thinking of. Rogue has the opposite problem, where
you had to be lucky enough that enough food even exists. At least in
my experience; I probably was really bad at rogue.

 

[Guest]

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

"Igor D. WonderLlama" <WonderLlama_@hotmail.com> wrote in
news:1121274378.438364.123840@f14g2000cwb.googlegroups.com:
> Billy Bissette writes:
>> "Igor D. WonderLlama" <WonderLlama_@hotmail.com> wrote in
>> news:1121207619.111233.72480@g49g2000cwa.googlegroups.com:
>>
>> Such as food and light generally are in Angband outside of Ironman
>> play. Sure, you can starve to death, but who really does outside of
>> constant paralysis?
>
> Exactly what I was thinking of. Rogue has the opposite problem, where
> you had to be lucky enough that enough food even exists. At least in
> my experience; I probably was really bad at rogue.

Rogue is also a really short game. Starving because you are out of
supplies 10 minutes into what? An hour or so game? is a bit different
than starving halfway through an Angband game. But the alternative is
pretty much to make food a non-issue.

 

[Guest]

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

On Tue, 12 Jul 2005 14:13:10 -0800, Courtney Campbell
<campbell@oook.cz> wrote:

>I believe people ignore the score because it doesn't provide any
>meaningful info. If it literally represented playing skill, people might
>use it to, say, compare their characters on a high score list.
>
>What exactly is on the list of things that defines skillful play?

There isn't any canonical list.

I don't think there is some ideal scoring method. Don't waste a lot of
time trying to invent one. I don't even think there is a "best"
scoring method, because any scoring system has to make assumptions
that won't hold across all players, games, character types, etc. And
roguelikes have a tradition of basing score on things that don't
really reflect skill very directly, rather than luck, things like how
much gold the PC has at death.

The only reason to even have a score is for the high score list, which
is really only interesting on a multi-user system.

A number of the things that define skillful play cannot be measured by
the score in a single game, because they manifest over multiple games.

Player A wins one game in twelve, taking an average of one million
turns to finish a winning game.

Player B wins three games in four, taking an average of two million
turns to finish a winning game.

Player A may well dominate the high score list if scoring is highly
influenced by turns taken, but is he necessarily the better player?

>Ah, yes, I feel much better now. I'd love some radical ideas on what
>'good play' is (so we can score it), and radical ideas on a time limit.

Outside the box thinking? Okay:

1. The ability to play with distractions. Therefore score should be
multiplied by 3 if the player has a spouse or SO living with him/her,
score should be multiplied by 2 for every dog and/or cat living
indoors and by 5 per child. Also, if you play with one hand while
masturbating to the internet porn you're downloading as you multitask,
score is multiplied by ten.

2. Use of spoilers should halve your score. Peeking at the source
should halve it again. So should asking questions on the newsgroup/a
mailing list/of passing strangers in the computer lab.

3. Have a real time clock on as well, used only for scoring purposes.

4. Destroying objects is just simplifying the dungeon, making it
require less skill to deal with. Penalize score every time the destroy
object command is used.

5. Don't provide a "score". List character level, win/death/still
living status, current or deepest (er, anti-deepest for Steam) dungeon
level, and list the ten most recent characters or ten most recent
winners.

--
R. Dan Henry
danhenry@inreach.com

 

[Guest]

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

On Tue, 12 Jul 2005 14:28:35 -0800, Courtney Campbell
<campbell@oook.cz> wrote:

>In article <1121181622.095055.326750@g14g2000cwa.googlegroups.com>,
> "Scott Yost" <yostage@gmail.com> wrote:
>
>> You could base the research time off of something else instead.
>> If you based it off of depth then it would actually be a finite
>> resource, and the player would have to make harder choices about what
>> to research. (maybe the research doesn't complete until the player
>> visits a depth that is current depth +2)
>
>I like depth as a finite resource, espically in Steamband where there
>are only 50 dungeon levels, but I don't think Steamware is the place for
>it.
>
>What are some other places depth could be used as a resource?
>-Campbell

It already is. Quests, you know.

--
R. Dan Henry
danhenry@inreach.com

 

[Guest]

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

On Tue, 12 Jul 2005 03:49:50 -0800, Courtney Campbell
<campbell@oook.cz> wrote:

>For those of you lucky enough to have tried Steamband, you may have
>noted I'm really trying to take it to the next level. (That's hyperbole)

(x^2)/4 - (y^2)/4 = 1

That's hyperbola.

>There are enough carbon copy variants of angband, and I'd like to make
>one that is a different play experience. Currently there is something
>called 'Steamware' which allows you to research and then install
>components into your 'body' giving you innate powers. The research takes
>money and a set amount of time, after which you can get it installed.
>
>Here is the deal. The 'set amount of time' is meaningless. Anyone can
>just buy it and then rest until it is ready. I see two solutions to the
>problem.

Is anybody actually doing that? I mean, if it's getting close to done,
I may hang out in town for awhile if I have something to do (been
playing aesthetes, so buying out the weapon shop, enchanting
everything, and selling it to fund my Steamware research works out).

Still, if you want something more, have the Steamware guy require you
go on a quest (for some rare material, book of lore, something
thematic) and create some new quest types for that (they could even be
separate from the main dungeon).

Honestly, for most characters, the financial costs mean waiting quite
awhile anyway. Waiting for the research too is a bit much.

--
R. Dan Henry
danhenry@inreach.com

 

[Guest]

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

On Tue, 12 Jul 2005 15:13:10 -0800, Courtney Campbell wrote:

> THERE IS NO REASON A VARIANT HAS TO PLAY, LOOK, OR ACT ANYTHING LIKE [V].

Ah, but you're posting on rec.games.roguelike.angband. It's reasonable to
assume that games posted on this group will play, look, and act like
Angband.

If your game will continue to depart from Angband as radically as ToME
has, then the time comes to move it to another newsgroup.

--
Neil Stevens - neil@hakubi.us

'A republic, if you can keep it.' -- Benjamin Franklin

 

[Guest]

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

Courtney Campbell <campbell@oook.cz> wrote:
>Eddie Grove <eddiegrove@hot.NOSPAM.mail.com> wrote:
>>For example, offer 99 !CCW in the temple, and that's it.
>>People who take too long run out.
>Sort of a reverse time limit? I sort of view it as six of one, half a
>dozen of the other. If the 'time limit' restricts you to 123 !CCW, or
>the store eventually runs out of them after 123 are sold, what is the
>quantifiable difference?

Some items are relatively common in the dungeon, and you might find enough
of them that you wouldn't be as eager to buy them unless you're having
problems with a certain unique or something like that. Rarer objects that
are commonly sold in the shops might skew this somewhat.


Otto Martin - grr, stupid player polymorph via radiation in Z, grr...
--
"You're splashing life into my eyes!"
http://countyoursheep.com/d/20041011.html

 

[Guest]

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

Neil Stevens <neil@hakubi.us> writes:

> On Tue, 12 Jul 2005 15:13:10 -0800, Courtney Campbell wrote:
>
>> THERE IS NO REASON A VARIANT HAS TO PLAY, LOOK, OR ACT ANYTHING LIKE [V].
>
> Ah, but you're posting on rec.games.roguelike.angband. It's reasonable to
> assume that games posted on this group will play, look, and act like
> Angband.
>
> If your game will continue to depart from Angband as radically as ToME
> has, then the time comes to move it to another newsgroup.

Which one? Steamband seems like a more radical departure than TOME: I
mean, the *stairs* go the wrong way!

cheers, Rich.



--
rich walker | Shadow Robot Company | rw@shadow.org.uk
technical director 251 Liverpool Road |
need a Hand? London N1 1LX | +UK 20 7700 2487
www.shadow.org.uk/products/newhand.shtml