Difficulty levels vs. save-scumming
Amy Wang
saying that everyone here should immediately implement it in their
games. After all, I've always said that features should be carefully
chosen by developers rather than taken blindly. I also understand that
the concept of difficulty levels isn't original in any sense, but I'm
bringing it up because I think it's relevant and hasn't been mentioned
in this form.
For my game (which isn't roguelike, but that doesn't really matter
right now), save-scumming doesn't occur because the game itself
doesn't encourage it. To many, Anubis seems to be draconian because
he's designing his game to encourage save-scumming, while at the same
time going to great lengths to artificially prevent it.
Save-scumming can be discouraged, from within the game, with many
mechanisms. For example, having a short game time would discourage it.
Spending ten hours or more playing a game, having the PC die, and
having nothing to show for it is very discouraging. The most natural,
and most difficult, way of discouraging save-scumming is to make sure
that the player has fun playing the game regardless of the outcome. I
won't offer any solutions to this, because I don't have any. I will
say that repeatedly playing the early-game with a weak-ass character
with no special weapons or abilities doesn't look nearly as fun when
described that way.
What I'm getting at here, however, is the concept of having difficulty
levels. Let's keep in mind that every game has at least one difficulty
level, so the concept isn't really that alien. Furthermore, unless
something really exotic is used for determining difficulty, the number
of possible levels isn't as limited as in commercial games (where
levels like 'hard', 'medium', and 'easy' are generally chosen simply
for ease of selection).
The obvious question is: why is this beneficial? Well, in a perfect
world, we don't want players to use save-scumming, and we've selected
permadeath as a feature (the discussion of save-scumming isn't really
relevant to games without permadeath, after all). So, instead of
forcing an unskilled player to either save-scum or replay the same
beginning fraction of the game, the idea is to lower the bar enough
for these players to be able to appreciate the whole of the game.
Easier levels let casual players play a complete game, while harder
levels allow hardcore gamers to gloat on Usenet about their
accomplishments.
Of course, the concept of difficulty levels won't be attractive to
many of you, for reasons that I'm already aware of. Most importantly,
many roguelike games are designed under the principle that features
should be withheld from players. Just making it to the elemental
planes in Nethack is an extreme accomplishment. Many PCs are lost
there, or even at the astral plane, but it is felt that the players
can more greatly appreciate getting to the end-game if it happens
rarely. However, if you design with the philosophy that the players
would have more fun experiencing all of the features of the game, ways
of encouraging that should be looked at. Neither approach is right or
wrong, but I think that each should be considered.
I wouldn't have mentioned this if I hadn't thought it was easy to
implement. Perhaps the most idiot-proof method is to change monster HP
levels. At the beginning of the game, the player could be presented
with a prompt for "select monster HP (%)". Then, in a generate_monster
function, the monster's hitpoints are adjusted in relation to this
number. Of course, this still leaves the PC equally open to damage,
but the concept of a difficulty level still holds. I'm sure this group
has many more elegant ways of handling difficulty.
I'm interested in whatever thoughts people have on this, but please
spare us the apologetics. I know that Nethack's classes have different
difficulties associated with them. That's just poor game balance, not
a real feature. Descending down levels doesn't count as game balance.
It's the way the game is played, not an option. Different Angband
variants may have different difficulties, but it's not widely
documented which is which, and you know that isn't the point. I know
that some of the people here are capable of intelligent responses.
Let's see them.
someog
"Amy Wang" <blueme...@hotmail.com> wrote in message
news:62acd54d.0301...@posting.google.com...
Hmm...it could work. The simplest idea that occurs to me is to allow
players an intrinsic AC bonus, which would result in near-immunity if chosen
high. That way, they'd have some recourse against off-screen instadeath by
poison breath. Or just divide all damage received by a scaling factor, to
reduce it if desired.
However, I confess I never really understood the practice of save-scumming
(though I'm not trying to resurrect the thread discussing it). When I first
played Angband, at v1.3.4, I save-scummed, and beat the game a few times.
Looking back, was it really worth it to invest long, long hours, with the
aid of a few batch files, to see an ASCII-graphic of a crown, and a
congratulation that I didn't earn? The gameplay is everything, the ending
almost nothing. People don't care about winning, they care about being able
to say they won. If achieved by save-scumming, then the game hasn't really
been beaten, because save-scumming effectively prevents the player from ever
dying.
So what I'm taking way too long to say, is why not just cheat death?
-David
be...@sonic.net
>
> Disclaimer: I'm not putting this forward as *the* solution.
> The most natural, and most difficult, way of discouraging
> save-scumming is to make sure that the player has fun playing
> the game regardless of the outcome.
> with no special weapons or abilities doesn't look nearly as fun
> What I'm getting at here, however, is the concept of having difficulty
> levels.
> the idea is to lower the bar enough
> for these players to be able to appreciate the whole of the game.
> Easier levels let casual players play a complete game, while harder
> levels allow hardcore gamers to gloat on Usenet about their
> accomplishments.
> many roguelike games are designed under the principle that features
> should be withheld from players.
I've clipped a lot of what you wrote, but I do appreciate it;
it was generally well-spoken and it's not a bad idea. I guess
the question here is how important is it to the game to preserve
its secrets? Is the player going to be as involved or into it
when he makes it to the elemental planes on the "hard" setting
if he's done it before on the "easy" and "medium" settings and
already knows what everything is and how to respond to everything?
The debate about savescumming, or difficulty levels, is all about
player involvement. The question is how you keep the player
involved with the character, and the game having new and unfamiliar
challenges at each new level is the primary answer. The prevention
of savescumming reserves surprises (and therefore involvement) for
the player at higher levels. Difficulty levels present an easier
challenge for novices, but at the expense of there being any
unfamiliar challenges left when they are playing a full-on game
and get to the higher levels.
I agree with you though, that playing a weak character through
repetitions of the earliest levels for the nth time is boring,
and doesn't add to the game's enjoyment. That was my main
reason for trying to do a roguelike; I wanted to make it different
enough each time, by using different monsters, variants, and
treasures, that the early levels could still have surprises
for you when you got killed and started over, at least if you
started over in a different game. But that's hard to implement
and hard to balance, and not even appropriate for a lot of
games.
Difficulty levels is one option; another is to allow the player
to cheat death, but at some severe penalty (wakes up at home, but
loses half his levels, all his equipment, and has all subsequent
experience cut by a third or something). That gives you something
in between the "Perfect" win where you did the whole thing
without dying and the "scum" win where you died a lot but weren't
affected at all by dying. It withholds the upper levels for people
who actually do well, either by advancing in spite of the penalty
or by not coming under the penalty, and simultaneously allows
players to avoid constantly repeating the very earliest part of
the game.
Hmmm, I could build that whole mechanic into a "Death-cheating
amulet," and have it around as a magic item. Of course, using it
once would lose it along with all the rest of your stuff. It's
even mildly plausible that some crackpot wizard would resurrect
people (albeit somewhat diminished in life-energy) in exchange for
a nice selection of magic items, so might make a few of these
things and scatter them around to garner himself all the cool
gear they'd bring in.
Bear
The Sheep
> Spending ten hours or more playing a game, having the PC die, and
> having nothing to show for it is very discouraging. The most natural,
> and most difficult, way of discouraging save-scumming is to make sure
> that the player has fun playing the game regardless of the outcome. I
> won't offer any solutions to this, because I don't have any. I will
> say that repeatedly playing the early-game with a weak-ass character
> with no special weapons or abilities doesn't look nearly as fun when
> described that way.
I never thought of it this way, but I think you're right.
Good old Rogue was fun to play even with low-level character,
because the items you found were random. You could find powerful
scrolls and potions on the first level.
I think Nethack has got similar advantages.
You could make it fun by providing many kinds of interesting
low-level monsters (those with special attacks, corpses with
eating-effects, etc.) and items.
The high-level monsters and items would be more powerful, but
with few new features.
This way playing from the beginning may be even more interesting
than continuing with powerful character -- but I think the challenge
can recompensate this.
And another plus -- debugging may be easier if all the features may
be encountered at early stages.
--
The Sheep
In search of real world:
cd /..
Taichen
>Save-scumming can be discouraged, from within the game, with many
>mechanisms. For example, having a short game time would discourage it.
>Spending ten hours or more playing a game, having the PC die, and
>having nothing to show for it is very discouraging. The most natural,
>and most difficult, way of discouraging save-scumming is to make sure
>that the player has fun playing the game regardless of the outcome. I
I think the solution for this is the "never ending" game someone
proposed not long ago ... you continue as a ghost, vampire, or go to
heaven / hell or whatever ... I think i' ll try to have this in my
future RL ..
Kornel "Anubis" Kisielewicz
news:62acd54d.0301...@posting.google.com...
> For my game (which isn't roguelike, but that doesn't really matter
> right now), save-scumming doesn't occur because the game itself
> doesn't encourage it. To many, Anubis seems to be draconian because
> he's designing his game to encourage save-scumming, while at the
same
> time going to great lengths to artificially prevent it.
Encourage save-scumming??? Oh, I get it...
> Save-scumming can be discouraged, from within the game, with many
> mechanisms. For example, having a short game time would discourage
it.
Yes, in my case though this is out of the picture :(.
> Spending ten hours or more playing a game, having the PC die, and
> having nothing to show for it is very discouraging. The most
natural,
> and most difficult, way of discouraging save-scumming is to make
sure
> that the player has fun playing the game regardless of the outcome.
I
> won't offer any solutions to this, because I don't have any. I will
> say that repeatedly playing the early-game with a weak-ass character
> with no special weapons or abilities doesn't look nearly as fun when
> described that way.
I have a few ideas to that though:
1) make the player remembered -- let someone in the game world mention
him
2) add some continuity between characters (there was a project once
called Karma, does anyone remember?)
3) make the player interested in playing another game (high
randomisation)
4) resign from powerlevelling (your 20 hour character that died isn't
much powerful then a new one) -- that needs careful balancing
though -- the player needs some other things that are worthy playing :
eg. plot.
5) and more...
> The obvious question is: why is this beneficial? Well, in a perfect
> world, we don't want players to use save-scumming, and we've
selected
> permadeath as a feature (the discussion of save-scumming isn't
really
> relevant to games without permadeath, after all). So, instead of
> forcing an unskilled player to either save-scum or replay the same
> beginning fraction of the game, the idea is to lower the bar enough
> for these players to be able to appreciate the whole of the game.
That's assuming that the beginning of the game will always be the
same -- that can be prevented by high randomness (I also count on
that).
> Easier levels let casual players play a complete game, while harder
> levels allow hardcore gamers to gloat on Usenet about their
> accomplishments.
Well... hmmm In the case of GenRogue this seperation is done
gamewise -- If you don't feel neither daredevil, nor realy good, you
can play an easy game, by avoiding all hard missions/places/towns etc.
I think that's a good approach -- and most cRPG games don't have itm
because they have fixed plots and goals...
regards,
--
Kornel "Anubis" Kisielewicz
RLDev Code v.0.65
L:FP E+ T+ R+++ P+ D++ G++ RL-- RLA+++ F:GearHead
GenRogue Reloaded ( http://genrogue.felis7.civ.pl/ )
W:DF Q+++ AI++ !GFX !SFX RN+++ PO--- Hp-- Re+++ S+++
Björn Bergström
news:62acd54d.0301...@posting.google.com...
[snip interesting stuff]
> I'm interested in whatever thoughts people have on this, but please
> spare us the apologetics. I know that Nethack's classes have different
> difficulties associated with them. That's just poor game balance, not
> a real feature. Descending down levels doesn't count as game balance.
> It's the way the game is played, not an option. Different Angband
> variants may have different difficulties, but it's not widely
> documented which is which, and you know that isn't the point. I know
> that some of the people here are capable of intelligent responses.
> Let's see them.
I agree with Amy. Changing the difficulty level of the game to prevent
save-scumming is a good solution. Up until recently I've used save scumming
when playing RLs. I stopped when trying Crawl. I really like the perma death
of RLs, but it isn't that fun to die with your high level character because
of a single stupid mistake you made... Amy's idea with difficulty levels is
a really good one and it can be implemented in many different ways:
o reduce monster HP
o increase player HP
o give player more starting xp for skills
o give player better starting equipment
o some kind of respawn feature where you're resurrected at the last altar
that you sacrificed permanent MP/HP/whatever at
Anyways, good point Amy! Can I reproduce the article at my webpage?
--
Björn Bergström
L:C++ E+ T- R+ P+ D-- G+ F:V RL-- RLA++
W:F Q+++ AI++ GFX+ !SFX RN+++ PO+ Hp- Re+ S++
Dungeondweller (http://roguelikedevelopment.org/dungeondweller/)
Roguelike Development (http://roguelikedevelopment.org)
Amy Wang
> I've clipped a lot of what you wrote, but I do appreciate it;
> it was generally well-spoken and it's not a bad idea. I guess
> the question here is how important is it to the game to preserve
> its secrets? Is the player going to be as involved or into it
> when he makes it to the elemental planes on the "hard" setting
> if he's done it before on the "easy" and "medium" settings and
> already knows what everything is and how to respond to everything?
Perhaps, difficulty levels wouldn't be the best solution for Nethack
(although I'm tempted to give it a try). In terms of preserving
secrets, the choice is left up to the developer's and players'
priorities. I would say that in the case of Nethack, it really isn't
as vital as it could be assumed. Spoiler files make sure of that,
because they are used by nearly all 'serious' roguelike gamers.
This does give me a reason to expand my thesis, though. With
difficulty levels, save-scumming *will not* occur, at least not in the
general sense. Given the choice between save-scumming or starting over
at a lower difficulty level, it seems to me that a player would prefer
to play through an easier game than awkwardly save/load through the
harder game. Furthermore, spoiler files wouldn't be used as much,
either. Even if I can plow through the challenges the game presents,
it would be more fun to figure out the features for myself than read
the research of someone else.
>
> The debate about savescumming, or difficulty levels, is all about
> player involvement. The question is how you keep the player
> involved with the character, and the game having new and unfamiliar
> challenges at each new level is the primary answer. The prevention
> of savescumming reserves surprises (and therefore involvement) for
> the player at higher levels. Difficulty levels present an easier
> challenge for novices, but at the expense of there being any
> unfamiliar challenges left when they are playing a full-on game
> and get to the higher levels.
I suppose the most relevant variable is the player's mentality. If the
player depends on discovering secrets, difficulty levels wouldn't make
the game more fun (except when the player seems to reach a plateau). I
believe, however, that there is a (perhaps larger) group of players
who care more about the plot of the game. In dungeon crawl roguelikes,
the plot tends to be nonexistant until the end game. Therefore, these
players will use whatever mechanisms are at their disposal (such as
save-scumming) to reach the end game.
The core of the issue is probably replayability. Important to all
games, replayability is even more important to roguelike games (for
reasons I'm not really aware of, but willing to accept). If players
can have fun playing the game even after winning it (which doesn't
seem too unlikely, since roguelikes have a lot of randomized things)
then replay value isn't hurt. Of course, it's the developer's choice
of where he wants the PC death curve. Any approach could be
legitimate, so long as some group of players would prefer it over an
alternative. It's only inappropriate to assume out-of-hand that all
players would prefer to play through 10% of the game a thousand times
than 100% a hundred times.
>
> I agree with you though, that playing a weak character through
> repetitions of the earliest levels for the nth time is boring,
> and doesn't add to the game's enjoyment. That was my main
> reason for trying to do a roguelike; I wanted to make it different
> enough each time, by using different monsters, variants, and
> treasures, that the early levels could still have surprises
> for you when you got killed and started over, at least if you
> started over in a different game. But that's hard to implement
> and hard to balance, and not even appropriate for a lot of
> games.
As a developer, you aren't responsible for making "a lot of games".
One game is enough, which gives you much more freedom than the
restraints of 'perfection' that's touted here. You don't need a
solution that would work for every roguelike game ever. You have the
choice of any world-specific features that you like, even if such
features would be foolish in another game world.
>
> Difficulty levels is one option; another is to allow the player
> to cheat death, but at some severe penalty (wakes up at home, but
> loses half his levels, all his equipment, and has all subsequent
> experience cut by a third or something). That gives you something
> in between the "Perfect" win where you did the whole thing
> without dying and the "scum" win where you died a lot but weren't
> affected at all by dying. It withholds the upper levels for people
> who actually do well, either by advancing in spite of the penalty
> or by not coming under the penalty, and simultaneously allows
> players to avoid constantly repeating the very earliest part of
> the game.
That's a great idea. After all, even expert players can run into bad
luck. Telling these players to start over encourages save-scumming,
but letting them continue with crippled characters gives them at least
a chance at whatever challenge killed the PC. Even if the PC dies 100
turns later, it could be very satisfying to defeat something that's
become proportionately more dangerous (implying that the player used
his wits to overcome it).
>
> Hmmm, I could build that whole mechanic into a "Death-cheating
> amulet," and have it around as a magic item. Of course, using it
> once would lose it along with all the rest of your stuff. It's
> even mildly plausible that some crackpot wizard would resurrect
> people (albeit somewhat diminished in life-energy) in exchange for
> a nice selection of magic items, so might make a few of these
> things and scatter them around to garner himself all the cool
> gear they'd bring in.
Yes, besides general mechanics that are usually the topics of
discussion, which are usually pretty limited, there's infinite
possibilities for game-specific mechanics that could be used, such as
what you've mentioned. A couple of additional examples immediately pop
into my mind. We could say that on the world map, each place (such as
a town) has a sphere of influence. If the PC dies close to a town, he
wakes up in its hospital. Near to a hermit's encampment, the hermit
could offer a less perfect cure. As for another (more game-specific)
example, we could say that when the PC dies, you immediately control a
ghost. This ghost could have a bunch of weaknesses (such as not being
able to use items), but it would give you a chance to find a
necromancer or alchemist to attempt to get started again.
Amy Wang
>
> Hmm...it could work. The simplest idea that occurs to me is to allow
> players an intrinsic AC bonus, which would result in near-immunity if chosen
> high. That way, they'd have some recourse against off-screen instadeath by
> poison breath. Or just divide all damage received by a scaling factor, to
> reduce it if desired.
Those are both good ways of doing it, I think.
>
> However, I confess I never really understood the practice of save-scumming
> (though I'm not trying to resurrect the thread discussing it). When I first
> played Angband, at v1.3.4, I save-scummed, and beat the game a few times.
> Looking back, was it really worth it to invest long, long hours, with the
> aid of a few batch files, to see an ASCII-graphic of a crown, and a
> congratulation that I didn't earn? The gameplay is everything, the ending
> almost nothing. People don't care about winning, they care about being able
> to say they won. If achieved by save-scumming, then the game hasn't really
> been beaten, because save-scumming effectively prevents the player from ever
> dying.
>
> So what I'm taking way too long to say, is why not just cheat death?
On the other hand, would it have been worth it to spend the same
number of hours and *not* win? Angband encourages save-scumming, even
if its developer didn't intend it to. You're given no real reason to
play after winning, so the game has replay value in its ability to
prevent players from winning. There's no plot to appreciate, or even
the special features found in other roguelikes.
Now, don't think that I'm trashing Angband. If it were an inferior
game, nobody would play it. It has an appeal to players who appreciate
a narrow (monsters, items, and maps) approach to replayability. But,
for a goal-oriented player, who expects some kind of special event to
happen when he wins, there really isn't any point, as you've
discovered. Such players will do whatever they can to reach that win
condition, while people who play the game the 'right way' understand
that there really isn't much of a point in winning, so they won't go
through extra effort to do so.
Some players will save-scum because the value of the game lies in the
entire game. They want to see all of its features, but the game itself
prevents them from doing that. It's simply human nature to struggle
against such an obstacle, and when pushed against a wall, people like
Julian will fight back. When this obstacle is removed, there ceases to
be any reason for save-scumming. We don't like save-scumming because
it represents the frustration of the player. Artificially preventing
it (with anti-save-scumming schemes) cures the symptom but not the
problem.
Amy Wang
[snip]
>
> You could make it fun by providing many kinds of interesting
> low-level monsters (those with special attacks, corpses with
> eating-effects, etc.) and items.
>
> The high-level monsters and items would be more powerful, but
> with few new features.
>
> This way playing from the beginning may be even more interesting
> than continuing with powerful character -- but I think the challenge
> can recompensate this.
>
> And another plus -- debugging may be easier if all the features may
> be encountered at early stages.
That's an excellent idea. Fighting rats and newts isn't fun. Goblins
in Nethack are marginally more fun because they might have a dagger,
but that isn't very satisfying. Crawl gives some starting monsters
poison needles and blowguns, which adds some interest, but often adds
unfair difficulty. There are many ways of adding features without
making the game intrinsically harder. Off-hand, I don't have any
easy-to-implement ways that you haven't already listed, but those are
viable.
Eytan Zweig
> > player involvement. The question is how you keep the player
> > involved with the character, and the game having new and unfamiliar
> > challenges at each new level is the primary answer. The prevention
> > of savescumming reserves surprises (and therefore involvement) for
> > the player at higher levels. Difficulty levels present an easier
> > challenge for novices, but at the expense of there being any
> > unfamiliar challenges left when they are playing a full-on game
> > and get to the higher levels.
>
> I suppose the most relevant variable is the player's mentality. If the
> player depends on discovering secrets, difficulty levels wouldn't make
> the game more fun (except when the player seems to reach a plateau). I
> believe, however, that there is a (perhaps larger) group of players
> who care more about the plot of the game. In dungeon crawl roguelikes,
> the plot tends to be nonexistant until the end game. Therefore, these
> players will use whatever mechanisms are at their disposal (such as
> save-scumming) to reach the end game.
>
I don't think you're right about plot; at least for general player
mentality. For me, *if* there is a plot, I will save-scum, because then my
mentality is "I want to advance the plot/see what happens next". If there is
no plot, I will not save scum, because then I'm thinking "I want to win".
> > Difficulty levels is one option; another is to allow the player
> > to cheat death, but at some severe penalty (wakes up at home, but
> > loses half his levels, all his equipment, and has all subsequent
> > experience cut by a third or something). That gives you something
> > in between the "Perfect" win where you did the whole thing
> > without dying and the "scum" win where you died a lot but weren't
> > affected at all by dying. It withholds the upper levels for people
> > who actually do well, either by advancing in spite of the penalty
> > or by not coming under the penalty, and simultaneously allows
> > players to avoid constantly repeating the very earliest part of
> > the game.
>
> That's a great idea. After all, even expert players can run into bad
> luck. Telling these players to start over encourages save-scumming,
> but letting them continue with crippled characters gives them at least
> a chance at whatever challenge killed the PC. Even if the PC dies 100
> turns later, it could be very satisfying to defeat something that's
> become proportionately more dangerous (implying that the player used
> his wits to overcome it).
The problem is, people will still save-scum. Again, using myself as an
example, I am far more likely to save-scum if there's a penalty than if the
game fully ends. If the game ends, I'm thinking "Game over. Time to
restart". If the game allows me to continue with a penalty, I think "that
sucks; lets try over".
Anyway, my point is that you shouldn't try to second-guess players. Some
tend to do one thing, others will tend to do another. All of your
suggestions are good - but not for preventing cheating. They're good for
providing the game with atmosphere and style. Whether people cheat or not
depends on the people in question, not on the game.
Eytan
Julian Mensch
Exactly. She gets it.
This is exactly what I'm doing in my own game, as
a matter of fact. I will post more details on my
plans as time permits.
Best regards,
-- Julian Mensch
Amy Wang
> I agree with Amy. Changing the difficulty level of the game to prevent
> save-scumming is a good solution. Up until recently I've used save scumming
> when playing RLs. I stopped when trying Crawl. I really like the perma death
On the topic of Crawl, it seems to me that 1st level characters are
overly-frail. If I can get a character to level three, it usually
makes it to ten or more. It's really discouraging to be unable to play
with several race/class combinations that start out weak.
> of RLs, but it isn't that fun to die with your high level character because
> of a single stupid mistake you made... Amy's idea with difficulty levels is
> a really good one and it can be implemented in many different ways:
>
> o reduce monster HP
> o increase player HP
> o give player more starting xp for skills
> o give player better starting equipment
> o some kind of respawn feature where you're resurrected at the last altar
> that you sacrificed permanent MP/HP/whatever at
These all seem like good ways of increasing player enjoyment. I
wouldn't have thought of giving the PC better equipment. Although it
is a one-time benefit, the potential of giving the player a decent
chance at experiencing the game is attractive.
>
> Anyways, good point Amy! Can I reproduce the article at my webpage?
Go ahead. I'm looking forward to seeing a new generation of roguelike
games.
Björn Bergström
> "Bj rn Bergstr m" <bjorn.b...@roguelikedevelopment.org> wrote in
message news:<b0739a$meib7$1...@ID-134817.news.dfncis.de>...
> > I agree with Amy. Changing the difficulty level of the game to prevent
> > save-scumming is a good solution. Up until recently I've used save
scumming
> > when playing RLs. I stopped when trying Crawl. I really like the perma
death
>
> On the topic of Crawl, it seems to me that 1st level characters are
> overly-frail. If I can get a character to level three, it usually
> makes it to ten or more. It's really discouraging to be unable to play
> with several race/class combinations that start out weak.
>
It seems like the beta 25 is much more difficult than beta 24. I haven't
player b25 yet, but I've seen many posts to r.g.r.m where people complain
that there characters die all the time.
[snip]
>
> >
> > Anyways, good point Amy! Can I reproduce the article at my webpage?
>
> Go ahead. I'm looking forward to seeing a new generation of roguelike
> games.
Thanks! BTW, what kind of game are you working on yourself? I read that it
wasn't a RL, am I right?
The Sheep
> That's an excellent idea. Fighting rats and newts isn't fun. Goblins
> in Nethack are marginally more fun because they might have a dagger,
> but that isn't very satisfying.
Quite some time ago, someone posted on the polish roguelike newsgroup
a description of her's shortest game:
It was something like this:
welcome to the dungeons of...
a goblin picks up a wand
goblin zaps a wand
death ray hits you
you die.
So even allowing monsters to pick up and use items adds a lot of variety.
someog
> "someog" <dsv...@yahoo.com> wrote in message
news:<b05b2p$lsnm5$1...@ID-175339.news.dfncis.de>...
> >
Yes, it would be worth it, if the reason I play (and it is) is to measure my
skills against the difficulty of the game. The computer games that I don't
care about are the ones I've beaten so many ways it isn't even fun to think
up new restrictions on myself. For me, Angband will never be in this
category, and so even when I lose, I enjoy it (after a brief
forehead-slapping interval). It's about the challenge, not bout seeing the
victory screen no matter the method.
Consider these ideas, and I doubt you'll disagree:
1) Save-scummers are either unable or unwilling to beat the game without
save-scumming.
2) Save-scummers who beat the game by so doing do not deserve the accolades
that come with winning.
3) Save-scummers who beat the game by so doing (for whatever reason), and
who do not ask for these accolades, are harmless. We should even accomodate
them if possible in their innocent diversions, since they harm no one.
4) Save-scumming protection schemes are generally built to thwart the
efforts of those in point 2.
5) There exists a mechanism by which those in point 3 can accomplish their
goal without save-scumming. In fact, they can do so with even less effort
than save-scumming requires. It's called cheating death, and it's already
available. The only thing they give up is what they didn't want anyway; the
ability to post their (invalid) scores.
So why fight those who implement save-scumming restrictions?
-David
Eytan Zweig
"someog" <som...@mailcity.com> wrote in message
news:b096u3$mk9tt$1...@ID-175339.news.dfncis.de...
That's true of some players, but hardly of all players. As I said in an
earlier message, I find plot to be a very strong encouragement to savescum -
because, unless plot is done really, really, well, I don't get immersed
enough to feel part of it; for 99% of games with good plots, I think "I'm
being told an interesting story" and then I try to get as much of the story
as possible, rather than play by the rules of the story. Savescumming is a
tool to see more of the plot.
Plotless games I treat, like David below, as an excersize. With no plot to
get in the way, all that matters is the game and the challange it offers.
There, savescumming serves no purpose.
Basically, I think that any game in which Amy will be encouraged to savescum
I won't be, and vice-versa. And there are probably hundrends of different
attitudes that other players may have. That's the real challange on people
who seek to prevent save-scumming by gameplay principles.
> Yes, it would be worth it, if the reason I play (and it is) is to measure
my
> skills against the difficulty of the game. The computer games that I
don't
> care about are the ones I've beaten so many ways it isn't even fun to
think
> up new restrictions on myself. For me, Angband will never be in this
> category, and so even when I lose, I enjoy it (after a brief
> forehead-slapping interval). It's about the challenge, not bout seeing
the
> victory screen no matter the method.
>
From here onwards, though I no longer agree with David's reasoning:
> Consider these ideas, and I doubt you'll disagree:
>
> 1) Save-scummers are either unable or unwilling to beat the game without
> save-scumming.
> 2) Save-scummers who beat the game by so doing do not deserve the
accolades
> that come with winning.
> 3) Save-scummers who beat the game by so doing (for whatever reason), and
> who do not ask for these accolades, are harmless. We should even
accomodate
> them if possible in their innocent diversions, since they harm no one.
> 4) Save-scumming protection schemes are generally built to thwart the
> efforts of those in point 2.
> 5) There exists a mechanism by which those in point 3 can accomplish their
> goal without save-scumming. In fact, they can do so with even less effort
> than save-scumming requires. It's called cheating death, and it's already
> available. The only thing they give up is what they didn't want anyway;
the
> ability to post their (invalid) scores.
>
> So why fight those who implement save-scumming restrictions?
>
You seem to be underestimating the power of save-scumming. Save-scumming
isn't just there to cheat death. Back when I used to save-scum, I scummed to
get better drops, I scummed to avoid minor effects such as stat drain, I
ID'd items by experimentation and scummed if I ended up causing bad effects,
and sometimes ID'd items by selling them and scummed if I got a bad deal.
Now, I was taking it to an extreme that most save-scummers probably don't,
but that doesn't mean some of them don't do part of that.
Note that I'm not arguing any particular point here, just pointing out that
the situation is more complex than presented. The entire issue of preventing
save-scumming seems to me fine in principle, but in practice it's a waste of
time and energy. Amy's difficulty setting idea, is a really good idea for
many reasons, but stopping save-scumming isn't one of them IMO.
Eytan
> -David
>
>
Kornel "Anubis" Kisielewicz
news:3pXV9.36$V36....@typhoon.nyu.edu...
> That's true of some players, but hardly of all players. As I said in
an
> earlier message, I find plot to be a very strong encouragement to
savescum -
> because, unless plot is done really, really, well, I don't get
immersed
> enough to feel part of it; for 99% of games with good plots, I think
"I'm
> being told an interesting story" and then I try to get as much of
the story
> as possible, rather than play by the rules of the story.
Savescumming is a
> tool to see more of the plot.
I would realy like to get your opinion on the way permadeath is used
in GenRogue -- I wrote about it in the Message-ID:
<b09k09$1lm$1...@news.tpi.pl> -- the second most recent reply to Julian
Mensch's post...
I would be realy happy, if you could find some spare time and reply to
that as you feel...
Especialy, that (succeding everywhere) != (seeing more of the plot) in
the case of GenRogue...
"Mårten
> "Amy Wang" <blueme...@hotmail.com> wrote in message
> news:62acd54d.03011...@posting.google.com...
>> "someog" <dsv...@yahoo.com> wrote in message
> news:<b05b2p$lsnm5$1...@ID-175339.news.dfncis.de>...
>> >
<snip>
> Consider these ideas, and I doubt you'll disagree:
>
> 1) Save-scummers are either unable or unwilling to beat the game without
> save-scumming.
> 2) Save-scummers who beat the game by so doing do not deserve the accolades
> that come with winning.
> 3) Save-scummers who beat the game by so doing (for whatever reason), and
> who do not ask for these accolades, are harmless. We should even accomodate
> them if possible in their innocent diversions, since they harm no one.
> 4) Save-scumming protection schemes are generally built to thwart the
> efforts of those in point 2.
> 5) There exists a mechanism by which those in point 3 can accomplish their
> goal without save-scumming. In fact, they can do so with even less effort
> than save-scumming requires. It's called cheating death, and it's already
> available. The only thing they give up is what they didn't want anyway; the
> ability to post their (invalid) scores.
I feel I'm in the #3 category.. but the #5 choice doesn't appeal to me at
all. When I play Angband/Zangband/Oangband I have no interest in posting
my possible wins to r.g.r.a (haven't actually won yet though, (and I've
been playing since 2.4-frog knows)) and the #5 choice just hinders my own
competition since I can't see if I got further with this caracter than the
last... Nobody but me cares if I win and nobody but me cares how I did it.
It's not until I start posting spoilers about plot etc that people will
start to care.
> So why fight those who implement save-scumming restrictions?
Why not?
In Final Fantasy X Me and a friend saved on one memory slot and tried some
things then we went back to that slot when we realised we made a misstake.
In fthe FF games TIME is realy important.. we don't have the time to play
a game that takes 50+ hours to complete just to realise when we've played
35 hours that we made a misstake in choosing X skill the first 5 hours and
that makes the end game almost uncompletable... That's how I feel about
roguelikes, I don't have the time to invest my feelings into 300
characters before I succed in reaching level 42. I'd like to be able to
play a character from start to end and if I happen to die and I did it on
a misstake on my account I will start over from the last save and learn
from that misstake. I should probably quit now since I'm quite drunk and
proably haven't spelled 50% correctly..
This is how I play games and has nothing to do with how the
programmers/designers intended it to be played. Checkout Morrowind for a
game that really lets you play the way you want.. I haven't completed it
yet because I think the main-plot is uninteresting at the moment and still
I have 40+ hours invested in the game...
> -David
Goodnight!
/Mårten
be...@sonic.net
>
> You seem to be underestimating the power of save-scumming. Save-scumming
> isn't just there to cheat death. Back when I used to save-scum, I scummed to
> get better drops, I scummed to avoid minor effects such as stat drain, I
> ID'd items by experimentation and scummed if I ended up causing bad effects,
> and sometimes ID'd items by selling them and scummed if I got a bad deal.
> Now, I was taking it to an extreme that most save-scummers probably don't,
> but that doesn't mean some of them don't do part of that.
>
> Note that I'm not arguing any particular point here, just pointing out that
> the situation is more complex than presented. The entire issue of preventing
> save-scumming seems to me fine in principle, but in practice it's a waste of
> time and energy. Amy's difficulty setting idea, is a really good idea for
> many reasons, but stopping save-scumming isn't one of them IMO.
I've observed this, actually. I played rogue way back in the wayback,
but the first time I saw Angband, one of my college roomies was playing
it. And he scummed in all the situations you mention and more. He'd
scum until he got the maximum hit point rolls when he went up a level -
every level. He scummed when he identified a potion of water so he
could
sell it unidentified instead. He scummed when he quaffed an unknown
potion
and it was good, if he felt that he might quaff it to better effect
later.
He'd even dig in the rubble, not find anything, and scum to dig in the
rubble again until he found something good.
I didn't say much, cause that was his style and it was his computer and
he'd downloaded the game fair and square. I felt like he had the right
to do whatever he wanted with it. But I observed, and there are some
mistakes in Angband that make it eminently scummable, which I'm not
going to repeat.
I've already fixed it so that if a particular character goes up a
particular level they get a particular number of hitpoints. You pick
your name, race, and class, and the name of the world, and your
hitpoint rolls at every level are foreordained from that point on.
Scumming will never produce a different result.
I'm still working on dungeon generation, but it's going to be fixed
so that all treasure drops are in place when the level is generated.
Scum all you want, but orc #345 on this level will always drop exactly
the same 12 copper coins, the chest in the corner will contain exactly
the same thing, and the dragon will have exactly the same stuff in its
hoard.
I figure it's futile to try to *prevent* savescumming, in the sense of
the character being able to change his/her actions and reap different
consequences. But on things that should *not* be affected by the
character's actions, at least the random number god shouldn't reward
scumming by allowing the player to pick among possible consequences.
Bear
Gerry Quinn
>I'm still working on dungeon generation, but it's going to be fixed
>so that all treasure drops are in place when the level is generated.
>Scum all you want, but orc #345 on this level will always drop exactly
>the same 12 copper coins, the chest in the corner will contain exactly
>the same thing, and the dragon will have exactly the same stuff in its
>hoard.
That leads to the inverse problem. If the left chest always contains a
bomb and the right chest always contains gold, cheaters will reload
after opening the wrong one.
- Gerry Quinn
Eytan Zweig
> I've observed this, actually. I played rogue way back in the wayback,
> but the first time I saw Angband, one of my college roomies was playing
> it. And he scummed in all the situations you mention and more. He'd
> scum until he got the maximum hit point rolls when he went up a level -
> every level. He scummed when he identified a potion of water so he
> could
> sell it unidentified instead. He scummed when he quaffed an unknown
> potion
> and it was good, if he felt that he might quaff it to better effect
> later.
> He'd even dig in the rubble, not find anything, and scum to dig in the
> rubble again until he found something good.
>
> I didn't say much, cause that was his style and it was his computer and
> he'd downloaded the game fair and square. I felt like he had the right
> to do whatever he wanted with it. But I observed, and there are some
> mistakes in Angband that make it eminently scummable, which I'm not
> going to repeat.
>
> I've already fixed it so that if a particular character goes up a
> particular level they get a particular number of hitpoints. You pick
> your name, race, and class, and the name of the world, and your
> hitpoint rolls at every level are foreordained from that point on.
> Scumming will never produce a different result.
>
Um, that's been the case in Angband as well, at least as far back as I'm
familiar with it (2.8.3). Also, the random seed is saved, so if you repeat
your next action exactly, you'll get the same result. Everything else you
mentioned is scummable in Angband, though.
Eytan
Eytan Zweig
"Kornel "Anubis" Kisielewicz" <anu...@felis7.civ.pl> wrote in message
news:b09rco$cjl$1...@news.tpi.pl...
> Użytkownik "Eytan Zweig" <ez...@nyu.edu> napisał w wiadomości
> news:3pXV9.36$V36....@typhoon.nyu.edu...
> > That's true of some players, but hardly of all players. As I said in
> an
> > earlier message, I find plot to be a very strong encouragement to
> savescum -
> > because, unless plot is done really, really, well, I don't get
> immersed
> > enough to feel part of it; for 99% of games with good plots, I think
> "I'm
> > being told an interesting story" and then I try to get as much of
> the story
> > as possible, rather than play by the rules of the story.
> Savescumming is a
> > tool to see more of the plot.
>
> I would realy like to get your opinion on the way permadeath is used
> in GenRogue -- I wrote about it in the Message-ID:
> <b09k09$1lm$1...@news.tpi.pl> -- the second most recent reply to Julian
> Mensch's post...
>
> I would be realy happy, if you could find some spare time and reply to
> that as you feel...
>
> Especialy, that (succeding everywhere) != (seeing more of the plot) in
> the case of GenRogue...
>
Well, I've read that post, and it seems very interesting - but what will
determine whether or not I will want to savescum in that game is not so much
how you design it, but how succesful you are in implementing it. The basic
issue is - if you're relying on immersiveness rather than a mental challange
as your game's primary motivation for success, first you need to be able to
get the players immersed - which is a hard task, and at best you'll succeed
for some people and fail for some since different people react to different
stuff. Then, you face the dilemma I mentioned above. Note that I'm not
saying I want to see all of the plot. I want to see the continuation of the
plot I've been experiencing.
Non-lethal failures, that keep the game interesting, are a great thing. If I
see that I can fail but still experience the game, I'll continue playing.
But any game-stopping failure means I have a choice between three options -
restarting, restoring a saved game and redoing, or quitting altogether. If
you eliminate the second of these choices, you're leaving me with a choice
between two options - and I'm not going to be more likely to restart, just
more likely to uninstall the game; that's because starting over is something
I very, very, very rarely do in a game with any semblance of plot. A game
like Angband, where I'm testing my skills, is always fun to retry. A story,
of any sort, is only fun to retry if it's a masterpiece. And even then, it's
worth experiencing mainly twice or three times, not dozens of times. There
are a *lot* of stories in this world - books to read, movies to watch, games
to play - and I want to experience as many of them as possible. Spending my
time replaying a game, as good as it may be, means I'm spending less time on
other stuff.
And note that turning the character into a wraith makes death a non-lethal
failure - though the game would have to remain interesting. If you cripple
the game once you die so that it's not fun anymore, that's even worse than
just shutting it off.
Experiencing the full consequences of my actions is, frankly enough, a
turn-off for me in a game. I have to experience the consequences of my
actions every day, all the time, in real-life. For a game, I go to relax and
escape from that. The reason why I think save/load features belong in games
which try to have any sort of realistic model of behavior in them is that
the pleasure for me in these games, other than experiencing an interesting
story, is that they offer me the chance to play around with consequences.
Note that I'm not trying to enter the permadeath/non-permadeath debate here,
or the savescumming/non-savescumming debate. Both permadeath and
non-permadeath are valid design choices in my book - heck, EyAngband has
permadeath - but given the rest of the design of GenRogue, permadeath means
that the rest of the game will have to be *VERY* good in order to maintain
my interest. Unless you manage to make me feel that starting a new character
and re-experiencing GenRogue will be worth my while more than any other form
of entertainment, it will be off to my "games I played in the past" pile the
moment my first serious character's game ends.
Oh, and the reason why I don't think you need implement any measures of
preventing save-scumming has nothing to do with gameplay philosphy, civil
rights, or computer safety. It's that any second you spend on preventing
savescumming is a second you're not spending on making the game *VERY* good.
Eytan
Paul Pekkarinen
> These all seem like good ways of increasing player enjoyment. I
> wouldn't have thought of giving the PC better equipment. Although it
> is a one-time benefit, the potential of giving the player a decent
> chance at experiencing the game is attractive.
Well this better equipment has already invented (with some other
benefits) in so called player profession or class. It's
easier to play Nethack with barbarian than tourist.
Who says we need difficulty levels like mentioned earlier?
And why roguelikes (or other games) should be made thinking those
wimps who cheat all the time in all games? "Pat pat, we'll make this
easy for you.. don't worry."
be...@sonic.net
It doesn't make me happy, but I'd rather have that than "if you try
opening it enough times it will contain exactly what you want."
Gerry Quinn
One answer might be to use random distributions, but keep a record.
If players have been excessively lucky, for whatever reason, a karmic
resonance causes chests to be less full of goodies than for those with
average luck...
Gerry Quinn
--
http://bindweed.com
Entertainment software for Windows
Puzzles, Strategy Games, Kaleidoscope Screensaver
Download evaluation versions free - no time limits
Morgoth's Cat
scribed:
>In article <3E2A4C11...@sonic.net>, be...@sonic.net wrote:
>>Gerry Quinn wrote:
>>>
>>> In article <3E28FF8B...@sonic.net>, be...@sonic.net wrote:
>>> >I'm still working on dungeon generation, but it's going to be fixed
>>> >so that all treasure drops are in place when the level is generated.
>>> >Scum all you want, but orc #345 on this level will always drop exactly
>>> >the same 12 copper coins, the chest in the corner will contain exactly
>>> >the same thing, and the dragon will have exactly the same stuff in its
>>> >hoard.
>>>
>>> That leads to the inverse problem. If the left chest always contains a
>>> bomb and the right chest always contains gold, cheaters will reload
>>> after opening the wrong one.
>>
>>It doesn't make me happy, but I'd rather have that than "if you try
>>opening it enough times it will contain exactly what you want."
>
>One answer might be to use random distributions, but keep a record.
>If players have been excessively lucky, for whatever reason, a karmic
>resonance causes chests to be less full of goodies than for those with
>average luck...
>
>Gerry Quinn
heh - I like that term. "Karmic Resonance". It has much possibilities.
How would it affect gameplay?
Best,
Dave
>--
>http://bindweed.com
>Entertainment software for Windows
>Puzzles, Strategy Games, Kaleidoscope Screensaver
>Download evaluation versions free - no time limits
E-mail: morgoth AT valinor DOT freeserve DOT co DOT uk
Homepage: http://www.veilofnight.net
Debunking Creationist Lies: The Supernovae and Supernova Remnants FAQ
http://www.talkorigins.org/faqs/supernova/
Visions of Light, Visions of Darkness - a B&W Photographic Gallery
http://www.valinor.freeserve.co.uk/visions.html
Gerry Quinn
>>One answer might be to use random distributions, but keep a record.
>>If players have been excessively lucky, for whatever reason, a karmic
>>resonance causes chests to be less full of goodies than for those with
>>average luck...
>
>How would it affect gameplay?
Not much except to discourage cheating, I hope. Players could try to
play a metagame of storing up bad luck in order to cheat when they
wanted without penalty, but a decent degree of algorithmic and
data obscurity could make this difficult.
- Gerry Quinn
David Damerell
>but the first time I saw Angband, one of my college roomies was playing
>it. And he scummed in all the situations you mention and more. He'd
>scum until he got the maximum hit point rolls when he went up a level -
>every level. He scummed when he identified a potion of water so he
>could sell it unidentified instead.
Scummed Wormtongue for Ringil, I hope?
>I didn't say much, cause that was his style and it was his computer and
>he'd downloaded the game fair and square. I felt like he had the right
>to do whatever he wanted with it.
But admit it... you thought it was incredibly lame?
>But I observed, and there are some
>mistakes in Angband that make it eminently scummable, which I'm not
>going to repeat.
I think it's pointless to deform your program to minimise the effects of
save-scumming. If people want to cheat, they will; even if Angband was
completely proof against save-scumming, pathetic people could edit the
lib/edit files (a useful feature which should not be removed) to make the
game easy.
>I've already fixed it so that if a particular character goes up a
>particular level they get a particular number of hitpoints.
Doesn't the current Angband do this?
>I'm still working on dungeon generation, but it's going to be fixed
>so that all treasure drops are in place when the level is generated.
>Scum all you want, but orc #345 on this level will always drop exactly
>the same 12 copper coins, the chest in the corner will contain exactly
>the same thing, and the dragon will have exactly the same stuff in its
>hoard.
Arguably this is a property of having monster inventories generated at
monster creation time, which leads one to monsters actually using all the
stuff they're carrying, which can only be a good thing.
--
David Damerell <dame...@chiark.greenend.org.uk> flcl?
David Damerell
>Well this better equipment has already invented (with some other
>benefits) in so called player profession or class. It's
>easier to play Nethack with barbarian than tourist.
>Who says we need difficulty levels like mentioned earlier?
Ah, but Amy doesn't like the different difficulties of different classes
in NetHack, even though they do, er, exactly what she wants. :-)
someog
news:PXi*0q...@news.chiark.greenend.org.uk...
> <be...@sonic.net> wrote:
> >but the first time I saw Angband, one of my college roomies was playing
> >it. And he scummed in all the situations you mention and more. He'd
> >scum until he got the maximum hit point rolls when he went up a level -
> >every level. He scummed when he identified a potion of water so he
> >could sell it unidentified instead.
>
CON and cLevel. Am I wrong?
David Damerell
>"David Damerell" <dame...@chiark.greenend.org.uk> wrote in message
>>>but the first time I saw Angband, one of my college roomies was playing
>>>it. And he scummed in all the situations you mention and more. He'd
>>>scum until he got the maximum hit point rolls when he went up a level -
>>>every level. He scummed when he identified a potion of water so he
>>>could sell it unidentified instead.
>Huh? I though the number of HP you got was a non-random function of your
>CON and cLevel. Am I wrong?
I didn't write any of that. Make your attribution lines appropriate to the
quoted text, please.
someog
Sorry.
-David
DarkGod
<som...@mailcity.com> wrote:
It doth have a random factor
--
-----------------------+----------------------------------------------
DarkGod comes from | Do not meddle in the affairs of wizards
the hells for YOU ! :) | because they are subtle and quick to anger.
-----------------------+----------------------------------------------
Pe W Olorin YSo L:50 DL:696 A+++ R+++ Sp++ w:Mage Staff of Mana(240%)
Pe*/PM*(Cr) D H- D c++ f- PV s- TT- d++ P++ M+ C- S++ I+++ So++ B/-
ac- GHB- SQ+ RQ V+++ F:Mage playing Mage-like(see Pernangband Sorcerors)
Eytan Zweig
"DarkGod" <dar...@ifrance.com> wrote in message
news:pan.2003.01.20....@ifrance.com...
> While under the effect of mushrooms of hallucination "someog"
> <som...@mailcity.com> wrote:
>
> > "David Damerell" <dame...@chiark.greenend.org.uk> wrote in message
> > news:PXi*0q...@news.chiark.greenend.org.uk...
> >> <be...@sonic.net> wrote:
> >> >but the first time I saw Angband, one of my college roomies was
playing
> >> >it. And he scummed in all the situations you mention and more. He'd
> >> >scum until he got the maximum hit point rolls when he went up a
level -
> >> >every level. He scummed when he identified a potion of water so he
> >> >could sell it unidentified instead.
> >>
> > Huh? I though the number of HP you got was a non-random function of
your
> > CON and cLevel. Am I wrong?
> It doth have a random factor
>
But it's a pre-determined random factor; it's impossible to alter it by
save-scumming.
Eytan
Greg McIntyre
> >But I observed, and there are some
> >mistakes in Angband that make it eminently scummable, which I'm not
> >going to repeat.
>
> I think it's pointless to deform your program to minimise the effects
> of save-scumming. If people want to cheat, they will; even if Angband
> was completely proof against save-scumming, pathetic people could edit
> the lib/edit files (a useful feature which should not be removed) to
> make the game easy.
I don't think adopting an approach different to Angband's constitutes
deforming your program (unless you're writing an Angband variant).
I save scum all the time in ADOM and it reduces satisfaction, but makes
the game more enjoyable overall because I dislike permadeath (although
I don't think it a bad thing in general).
In order not to diminish game satisfaction in my roguelike, and also to
handily make saving/loading the game far easier, I intend to follow the
design of Zelda games by allowing the player to save at any time
(anything less is annoying), however should they load the game back
again they'll be at the last 'safe' place they visited. Yes, you can
still save-scum, but damn it's annoying to trek back to try opening that
chest again.
People will argue that it's annoying in general to trek back to where
you left off, but it's much less annoying than permadeath, which is
an extreme case. ;)
--
Greg McIntyre
gr...@puyo.cjb.net
http://puyo.cjb.net
DarkGod
<ez...@nyu.edu> wrote:
>
> "DarkGod" <dar...@ifrance.com> wrote in message
> news:pan.2003.01.20....@ifrance.com...
>> While under the effect of mushrooms of hallucination "someog"
>> <som...@mailcity.com> wrote:
>>
>> > "David Damerell" <dame...@chiark.greenend.org.uk> wrote in message
>> > news:PXi*0q...@news.chiark.greenend.org.uk...
>> >> <be...@sonic.net> wrote:
>> >> >but the first time I saw Angband, one of my college roomies was
> playing
>> >> >it. And he scummed in all the situations you mention and more. He'd
>> >> >scum until he got the maximum hit point rolls when he went up a
> level -
>> >> >every level. He scummed when he identified a potion of water so he
>> >> >could sell it unidentified instead.
>> >>
>> > Huh? I though the number of HP you got was a non-random function of
> your
>> > CON and cLevel. Am I wrong?
>> It doth have a random factor
>>
>
> But it's a pre-determined random factor; it's impossible to alter it by
> save-scumming.
Never said it is :)
David Damerell
>David Damerell <dame...@chiark.greenend.org.uk> wrote:
>>>But I observed, and there are some
>>>mistakes in Angband that make it eminently scummable, which I'm not
>>>going to repeat.
>>I think it's pointless to deform your program to minimise the effects
>>of save-scumming.
>deforming your program (unless you're writing an Angband variant).
Uh, you know how much I dislike Angband; that wasn't what I meant. I'd be
a lot happier if people didn't seem so often to be carefully
reimplementing Angband.
What I meant was that the OP appeared to be making some design decisions
purely in order to prevent save-scumming; decisions that might even have
other negative consequences. I think that's a Bad Thing.
Greg McIntyre
> What I meant was that the OP appeared to be making some design
> decisions purely in order to prevent save-scumming; decisions that
> might even have other negative consequences. I think that's a Bad
> Thing.
Right, I'm sorry.
Jesse Welton
> [...] I guess
> the question here is how important is it to the game to preserve
> its secrets? Is the player going to be as involved or into it
> when he makes it to the elemental planes on the "hard" setting
> if he's done it before on the "easy" and "medium" settings and
> already knows what everything is and how to respond to everything?
One way to make "easy" levels easy is to actually remove some of the
challenges present at the harder levels. Cut out certain kinds of
nasty beasties, or cripple them so they can't use all their abilities.
Then players won't know how to respond to everything at the harder
levels.
-Jesse
"What the-- A *Death* Weeble?"
Jesse Welton
<gr...@puyo.cjb.net> wrote:
> In order not to diminish game satisfaction in my roguelike, and also to
> handily make saving/loading the game far easier, I intend to follow the
> design of Zelda games by allowing the player to save at any time
> (anything less is annoying), however should they load the game back
> again they'll be at the last 'safe' place they visited. Yes, you can
> still save-scum, but damn it's annoying to trek back to try opening that
> chest again.
This sounds extremely abusable. Can't escape the hoard of zombie monkeys
that are about to eat your brain? Simply save and restore, and you're
home free at the last safe point.
Dont forget, this isn't only annoying for save scummers. It affects
anyone who needs to suspend play, which means pretty much everyone who
plays your game. Why would you set out to make your save system
annoying for all your players?
-Jesse
Greg McIntyre
> In article <20030121142528...@puyo.cjb.net>, Greg McIntyre
> <gr...@puyo.cjb.net> wrote:
> > In order not to diminish game satisfaction in my roguelike, and also
> > to handily make saving/loading the game far easier, I intend to
> > follow the design of Zelda games by allowing the player to save at
> > game back again they'll be at the last 'safe' place they visited.
> > Yes, you can still save-scum, but damn it's annoying to trek back to
> > try opening that chest again.
>
> This sounds extremely abusable. Can't escape the hoard of zombie
> monkeys that are about to eat your brain? Simply save and restore,
> and you're home free at the last safe point.
Yup. That's intentional design. It doesn't let you escape the larger
problem of how to get past the hoard of zombie monkeys that lay in wait
on the next map (they refresh their numbers anew when you go back
there). It merely allows you to retreat and try again. My game is not a
permadeath affair, and that sort of behaviour is fine by me.
> Dont forget, this isn't only annoying for save scummers. It affects
> anyone who needs to suspend play, which means pretty much everyone who
> plays your game. Why would you set out to make your save system
> annoying for all your players?
Do you find the save/load system in Zelda games annoying? I don't.
That's because a vast majority of the time, I'm happy to play the game
until I reach a safe point to save, where I will be returned when I
restart the game. It's only when I urgently need to leave the game that
I'll save and exit knowing that I'll have to retrace my steps to get
back to where I currently am (e.g. I'm running late for uni, or my
sibling wants to use the computer for work). These are exceptional
circumstances, at least in my life.
Jesse Welton
<gr...@puyo.cjb.net> wrote:
> welt...@osu.edu (Jesse Welton) wrote:
> >
> > This sounds extremely abusable. Can't escape the hoard of zombie
> > monkeys that are about to eat your brain? Simply save and restore,
> > and you're home free at the last safe point.
>
> Yup. That's intentional design. It doesn't let you escape the larger
> problem of how to get past the hoard of zombie monkeys that lay in wait
> on the next map (they refresh their numbers anew when you go back
> there). It merely allows you to retreat and try again. My game is not a
> permadeath affair, and that sort of behaviour is fine by me.
Okay, so it sounds like the entire game state gets rolled back to the
way it was when you reached the safe point. Isn't this just equivalent
to auto-saving at each safe point, and not allowing saves elsewhere?
Anyway, that's not such an unbalancing feature if it rolls the whole
game back.
> Do you find the save/load system in Zelda games annoying? I don't.
I've never played a Zelda game, so I don't know the details of its save
system. However, I have played games with only discrete save points
available. IMO, such a system works best for action games, rather than
turn-based tactical games. The reason is that action games derive
enjoyment in large part from the development and exercise of skill with
the controls. Playing through the same section again isn't wasted time,
from that perspective. With turn-based tactical games like roguelikes,
it's more about making decisions to maximize one's chances against an
uncertain array of threats. I'll behave differently if I know the next
three rooms are empty, and the fourth contains a nest of phase gerbils,
than if I have only a vague notion of the kinds of pitfalls that I'm
likely to encounter at the current level. I wouldn't enjoy replaying
bits I've already been through. (That's precisely why the randomness
of roguelike games gives them such good replayability: you're never
just replaying the exact same bits as before; it's fresh every time.)
So, yes, I would find it annoying for saved games to make me replay
sections of a roguelike game.
> That's because a vast majority of the time, I'm happy to play the game
> until I reach a safe point to save, where I will be returned when I
> restart the game. [...]
Even when I'm not rushed, I often decide to save a roguelike game in
the middle of a dangerous situation, because that's when I need to be
more careful, considering my options at length upon returning to the
situation with a refreshed mind. Perhaps in the absence of permadeath
I wouldn't care as much about getting it right the first time.
Certainly, discrete save points help balance nonpermadeath action
games by requiring the player to at least complete an entire sequence
of the game in a single try.
-Jesse
Greg McIntyre
> Okay, so it sounds like the entire game state gets rolled back to the
> way it was when you reached the safe point. Isn't this just
> equivalent to auto-saving at each safe point, and not allowing saves
> elsewhere?
Not really. See below.
> Anyway, that's not such an unbalancing feature if it rolls
> the whole game back.
Mine is not a traditional roguelike! I know it makes you squirm and hate
me that I'm writing something that may more adequately be described as a
randomised, text-based, party-based, plot-heavy, Nintendo and old school
Squaresoft inspired CRPG. *G* It's a different *type* of game to your
average roguelike, so talking about balance assumes a whole lot which
doesn't really translate.
The entire game state is not rolled back. For example, each map can
define what it wants to save, the default being nothing. Meaning that if
the player spends a lot of time on a map which saves nothing, leaves and
comes back, the map has reset and appears exactly as it was when they
first arrived on it. Maps which save nothing, or very little, are the
ones with lots of bad guys that get placed between the player and the
rewards, plot points and other interesting stuff (on the next map).
> > Do you find the save/load system in Zelda games annoying? I don't.
>
> I've never played a Zelda game, so I don't know the details of its
> save system. However, I have played games with only discrete save
> points available. IMO, such a system works best for action games,
> rather than turn-based tactical games. The reason is that action
> games derive enjoyment in large part from the development and exercise
> of skill with the controls. Playing through the same section again
> isn't wasted time, from that perspective. With turn-based tactical
> games like roguelikes, it's more about making decisions to maximize
> one's chances against an uncertain array of threats.
You're correct, but mine is not a turn-based tactical game like a
roguelike. Mine is a turn-based tactical game more like... ah...
randomised, rogue-inspired, Final Fantasy where the fights are replaced
by encounters on a rogue-like overhead map. I'm probably to blame for
the confusion due to me calling it a roguelike and having others assume
I was taking Angband or Nethack as a template.
Roughly speaking, the game mechanic is such: you move from safe point to
safe point in the world, where interesting social interactions take
place. Between safe points there are sections of evil nasties out to
kill you (the world map is effectively a connected graph of maps). You
have to deal with the bad guys using turn-based tactics in order to
proceed on the main map. Conquered areas can be skipped over on the main
map, however you can still visit them if you like, to rejoice in
nostalgia or get more experience, or money, or just because you're a
sadist (or a masochist, as the case may be).
To avoid you kicking my arse because I raised this in a discussion about
save-scumming in more traditional roguelike games, I'd just like to say
that IMHO, the same Zelda-ish system could be implemented in more
traditional roguelikes and that people who care about players exploits
and realism should, proverbially speaking, get over it. :)
be...@sonic.net
>
> "Amy Wang" <blueme...@hotmail.com> wrote in message
> news:62acd54d.03011...@posting.google.com...
> > I suppose the most relevant variable is the player's mentality. If the
> > player depends on discovering secrets, difficulty levels wouldn't make
> > the game more fun (except when the player seems to reach a plateau). I
> > believe, however, that there is a (perhaps larger) group of players
> > who care more about the plot of the game. In dungeon crawl roguelikes,
> > the plot tends to be nonexistant until the end game. Therefore, these
> > players will use whatever mechanisms are at their disposal (such as
> > save-scumming) to reach the end game.
> I don't think you're right about plot; at least for general player
> mentality. For me, *if* there is a plot, I will save-scum, because then my
> mentality is "I want to advance the plot/see what happens next". If there is
> no plot, I will not save scum, because then I'm thinking "I want to win".
I have a question for you; if plot-capable elements are scattered
around the game, but the game designer has just thrown these elements
in more or less at random and left you to pick and choose among them
to sort out your own plot, do you still feel compelled to advance a
particular plot or see what happens next?
For example, you could be presented with a situation like this:
There is a war between the local tribe of goblins and the Empire of
the Orcs whose capitol is far away. Captain Squamish is trying to
move a caravan of goods across the hotly contested territory, and
unknown to him but known to you, the caravan contains a magic item
very important to the war. There's a band of barbarians hunting for
the magic item but they have no idea where to find it. The Orcish
empire is aware of its existence and may or may not know where it is.
The goblins are clueless about it but might attack the caravan looking
for food and/or slaves and/or mounts (not much difference to a
goblin, actually). And there's a band of robbers that are planning
an attack on the caravan. You're an Elf, you're in command of a
small force of rangers, and you are part of the Sidhe Combined Ranger
Force, whose ultimate goal is to wipe out both the goblin tribe and
the Empire of the Orcs because they're both cutting down sacred
forests your people depend on for their lives, magics, sacraments,
and livelihoods. You've just captured a goblin scout; your camp will
probably be discovered in a day or so, when his commander sends a
force to try to find out why he didn't return.
Now, you've got about fifty choices here; You could protect the
caravan so that neither side gets the magic item, tip off the
barbarians to the magic item's location so that they'll get it
instead of either of the warring monster tribes, or attack the
caravan yourself and try to take it. Or you could "join" the
robbers who are attacking the caravan and hope to grab the magic
item while they're busy splitting up the trade goods and ignorant
of its value, or claim it as your share of booty. You could use
one of your six remaining magic messenger arrows to send an
urgent letter back to your SCRF commander in the Sidhe Lands asking
for reinforcements. Or any of a buncha other things. Or you could
just walk up to captain Squamish, explain the situation, and ask
for the item. But.... The plot-capable elements are selected at
random. Your responses can only be a best guess, because the game
designer was focused on selecting and combining elements providing
an interesting situation with lots of possibilities, rather than
one with "one right answer." In fact, there's no guarantee that
*anything* you do in this situation will work, and it may just
be better to ignore it, break camp, and head west to see what
situation is developing, probably with respect to the same war,
three hills away.
Do you still feel compelled to savescum in order to "advance the plot"
when there is not a specific plot laid out for you and you're just
acting under best guesses? When any three of the above options, or
none, may lead to eventual victory if played out?
> Anyway, my point is that you shouldn't try to second-guess players. Some
> tend to do one thing, others will tend to do another. All of your
> suggestions are good - but not for preventing cheating. They're good for
> providing the game with atmosphere and style. Whether people cheat or not
> depends on the people in question, not on the game.
This is true, and I understand; but I'm just wondering what it is about
a "plot" that motivates you? Is it the feeling that it's there to be
uncovered/discovered, or would you feel the same if it were freestyle
and you knew that there wasn't a preselected "plot" and, indeed, might
not even be any particular "right answer"?
Bear
Eytan Zweig
<be...@sonic.net> wrote in message news:3E2F9741...@sonic.net...
> Eytan Zweig wrote:
> >
> > "Amy Wang" <blueme...@hotmail.com> wrote in message
> > news:62acd54d.03011...@posting.google.com...
>
> > > I suppose the most relevant variable is the player's mentality. If the
> > > player depends on discovering secrets, difficulty levels wouldn't make
> > > the game more fun (except when the player seems to reach a plateau). I
> > > believe, however, that there is a (perhaps larger) group of players
> > > who care more about the plot of the game. In dungeon crawl roguelikes,
> > > the plot tends to be nonexistant until the end game. Therefore, these
> > > players will use whatever mechanisms are at their disposal (such as
> > > save-scumming) to reach the end game.
>
> > I don't think you're right about plot; at least for general player
> > mentality. For me, *if* there is a plot, I will save-scum, because then
my
> > mentality is "I want to advance the plot/see what happens next". If
there is
> > no plot, I will not save scum, because then I'm thinking "I want to
win".
>
> I have a question for you; if plot-capable elements are scattered
> around the game, but the game designer has just thrown these elements
> in more or less at random and left you to pick and choose among them
> to sort out your own plot, do you still feel compelled to advance a
> particular plot or see what happens next?
[snip description]
> Do you still feel compelled to savescum in order to "advance the plot"
> when there is not a specific plot laid out for you and you're just
> acting under best guesses? When any three of the above options, or
> none, may lead to eventual victory if played out?
>
Well, I find it hard to answer this question in a general way - it depends a
lot on the indivudual game and how I feel about it. The short answer is, I
can't think offhand of any games which succesfully implement the sort of
plot you are describing. I'm not sure that if such a game existed, I'd
really be interested in playing it - I might be very impressed by it, but I
don't know if I'll enjoy it as much as a game with either a more-or-less
linear plot or one with no plot at all.
Of course, I may be wrong - the game might really appeal to me, and immerse
me, in which case I won't savescum. Or it might really appeal to me, but I'd
focus on some plot elements, and then save-scum in order to advance those at
the cost of others. But I don't think the design in itself would favour
either of these outcomes over the other.
>
> > Anyway, my point is that you shouldn't try to second-guess players. Some
> > tend to do one thing, others will tend to do another. All of your
> > suggestions are good - but not for preventing cheating. They're good for
> > providing the game with atmosphere and style. Whether people cheat or
not
> > depends on the people in question, not on the game.
>
> This is true, and I understand; but I'm just wondering what it is about
> a "plot" that motivates you? Is it the feeling that it's there to be
> uncovered/discovered, or would you feel the same if it were freestyle
> and you knew that there wasn't a preselected "plot" and, indeed, might
> not even be any particular "right answer"?
>
I don't know. I haven't yet seen a truely freeform plot that actually works
and isn't obviously fake and templatic. If I were to guess my reaction, I'd
assume that I'll find that interesting for a while but once real life will
force me to disassociate with it for even a short time (as it inevitably
will), I won't feel motivated to return.
Eytan
> Bear
Jesse Welton
<gr...@puyo.cjb.net> wrote:
>
> Mine is not a traditional roguelike! I know it makes you squirm and hate
> me that I'm writing something that may more adequately be described as a
> randomised, text-based, party-based, plot-heavy, Nintendo and old school
Not at all. In fact, it helps me understand your motivation for the
save system you propose. I still don't know that I agree with the
decision, though.
> To avoid you kicking my arse because I raised this in a discussion about
> save-scumming in more traditional roguelike games, I'd just like to say
> that IMHO, the same Zelda-ish system could be implemented in more
> traditional roguelikes and that people who care about players exploits
> and realism should, proverbially speaking, get over it. :)
This is supposed to make me *less* likely to want to shower you with
feet? :) Seriously, I wouldn't like this, for reasons already given
which have nothing to do with exploits or realism (though it suffers
on those points as well).
-Jesse
Kornel "Anubis" Kisielewicz
> > I would realy like to get your opinion on the way permadeath is
used
> > in GenRogue -- I wrote about it in the Message-ID:
> > <b09k09$1lm$1...@news.tpi.pl> -- the second most recent reply to
Julian
> > Mensch's post...
> >
> > I would be realy happy, if you could find some spare time and
reply to
> > that as you feel...
> >
> > Especialy, that (succeding everywhere) != (seeing more of the
plot) in
> > the case of GenRogue...
> >
>
> Well, I've read that post, and it seems very interesting - but what
will
> determine whether or not I will want to savescum in that game is not
so much
> how you design it, but how succesful you are in implementing it.
If I fail to do that, then the whole game wont have a point anyway ;).
> The basic
> issue is - if you're relying on immersiveness rather than a mental
challange
> as your game's primary motivation for success, first you need to be
able to
> get the players immersed - which is a hard task, and at best you'll
succeed
> for some people and fail for some since different people react to
different
> stuff.
I know. If I manage to immerse just about 5% of the potential gamers,
then I'll treat it as a success anyway.
> Then, you face the dilemma I mentioned above. Note that I'm not
> saying I want to see all of the plot. I want to see the continuation
of the
> plot I've been experiencing.
Note that what I've written -- the plot may become more interefting if
you also face failures...
> Non-lethal failures, that keep the game interesting, are a great
thing.
I know. But they're also hard to create, because of the branching
factor -- normal cRPG's usualy just concentrate on the success,
because they assume the player will Save/Load untill he succeeds (and
usualy either failure is fatal, or success doesn't have a great impact
on the plot...)
> If I
> see that I can fail but still experience the game, I'll continue
playing.
That's the main and hardest point...
> But any game-stopping failure means I have a choice between three
options -
> restarting, restoring a saved game and redoing, or quitting
altogether.
As I see it, the only game-stopping failure will be permanent death in
the underworld.
> you eliminate the second of these choices, you're leaving me with a
choice
> between two options - and I'm not going to be more likely to
restart, just
> more likely to uninstall the game; that's because starting over is
something
> I very, very, very rarely do in a game with any semblance of plot.
This is where the dynamical, random plot generation kicks in -- you
can have a new game, without the need of doing the same things twice.
> And note that turning the character into a wraith makes death a
non-lethal
> failure - though the game would have to remain interesting. If you
cripple
> the game once you die so that it's not fun anymore, that's even
worse than
> just shutting it off.
True. But as "Wraith : the Oblivion" showed, a life of a Wraith can be
truly interesting ;). And I will probably have one advantage over that
great RPG title -- the Wraith in my world will have a life that he
realy lived...
> Oh, and the reason why I don't think you need implement any measures
of
> preventing save-scumming has nothing to do with gameplay philosphy,
civil
> rights, or computer safety. It's that any second you spend on
preventing
> savescumming is a second you're not spending on making the game
*VERY* good.
Well, in a sense you're got a valid point here -- but
1 - the anti-ss mechanism is simple, and wont take much time,
2 - implementing the important features demads full concentration from
me, which I rarely can achieve. Implementing side things like anti-ss,
doesn't require that...
Anyway, thanks for your opinion on that Eytan. I realy appriciate
that. And truly regret the fact that I can't play EyAngband -- for
some unknown reason it hangs my whole system when executed (I use a
Toshiba Celeron 700 laptop running W98).... :-(
Eytan Zweig
> that. And truly regret the fact that I can't play EyAngband -- for
> some unknown reason it hangs my whole system when executed (I use a
> Toshiba Celeron 700 laptop running W98).... :-(
>
That's weird... Does that happen to other Angband variants as well?
Eytan
Kornel "Anubis" Kisielewicz
> > Anyway, thanks for your opinion on that Eytan. I realy appriciate
> > that. And truly regret the fact that I can't play EyAngband -- for
> > some unknown reason it hangs my whole system when executed (I use
a
> > Toshiba Celeron 700 laptop running W98).... :-(
> >
>
> That's weird... Does that happen to other Angband variants as well?
> Eytan
The same is true in my case with UnAngband -- the rest works fine
(tried GSN, GSN2, Pern, ToME 1,2, Vanilla, Norse, K, and a few
others....)
Amy Wang
> blueme...@hotmail.com (Amy Wang) wrote in message
> > These all seem like good ways of increasing player enjoyment. I
> > wouldn't have thought of giving the PC better equipment. Although it
> > is a one-time benefit, the potential of giving the player a decent
> > chance at experiencing the game is attractive.
>
> Well this better equipment has already invented (with some other
> benefits) in so called player profession or class. It's
> easier to play Nethack with barbarian than tourist.
Didn't I already explain why you're wrong? I guess I'll try again. The
barbarian is not equivalent to a wizard with lower difficulty. They
are different classes, implying a different way of playing the game.
The fact that the barbarian is 'easier' is merely due to the fact that
Nethack's developers spent no time on play balance, since the entire
point of Nethack is to have an unfair challenge. Why don't you stop
trying to dodge my point and argue against it if you don't like it?
> Who says we need difficulty levels like mentioned earlier?
I do. Several people here do. Every gamer in the entire world, with
the exception of a few 'hardcore' roguelike gamers, does. Like I said
before, however, I'm not saying that every game needs to have
difficulty levels. I'm merely saying that there are *many* cases in
which it would improve a game to have this feature.
> And why roguelikes (or other games) should be made thinking those
> wimps who cheat all the time in all games? "Pat pat, we'll make this
> easy for you.. don't worry."
That's not even a sentence. I'm not going to bother trying to
extrapolate some kind of meaning from it, either, since it is
undoubtedly just an adolescent attempt at diffusing an argument with
childish whining.
Amy Wang
>
> What I meant was that the OP appeared to be making some design decisions
> purely in order to prevent save-scumming; decisions that might even have
> other negative consequences. I think that's a Bad Thing.
As far as I'm concerned, save-scumming is just a symptom of the
problem. If a player is writing scripts or copying files around as
part of playing the game, it's showing a high level of
dissatisfaction.
Besides, you don't have to be such an idiot. Mainstream games have
difficulty levels. No roguelike game has them. What does that say?
Roguelike developers have been ignoring (and continue to ignore)
possible features that could improve their games.
If you're just going to dismiss a widely accepted feature as being a
'Bad Thing' due to some vague allusions to negative consequences, then
that's your arguement. I've put mine down already.
Amy Wang
> "Amy Wang" <blueme...@hotmail.com> skrev i meddelandet
> news:62acd54d.03011...@posting.google.com...
> > "Bj rn Bergstr m" <bjorn.b...@roguelikedevelopment.org> wrote in
> message news:<b0739a$meib7$1...@ID-134817.news.dfncis.de>...
> [snip]
> >
> > >
> > > Anyways, good point Amy! Can I reproduce the article at my webpage?
> >
> > Go ahead. I'm looking forward to seeing a new generation of roguelike
> > games.
>
> Thanks! BTW, what kind of game are you working on yourself? I read that it
> wasn't a RL, am I right?
I'm sorry to take so long to reply. I've been without internet for a
while.
I've explained my game's interface in other threads, but I'd be happy
to elaborate on relevant features. As a recap, it's a text-based,
mouse-driven game, in which the interface is designed around the
principle of being able to click on a game object (locations, people,
items, ...) to interact with it.
The entire issue of save-scumming and difficulty levels is avoided.
Currently, I haven't even implemented a save feature, since it isn't
really needed in the game. While a CRPG might go on for thirty hours,
and a roguelike game for just as long, my game can be played in
segments of an hour or two. I'm able to structure it that way because
of the horizontal distribution of features. Most of the functionality
of the game can be accessed within a half-hour. PCs don't gain levels
or experience in the sense of any RPG. Character 'progression' isn't
looked at as a design principle. Characters can change, but that
change isn't necessarily desirable. I might need to elaborate on that,
but I'll wait until it's solicited.
Since there's no win condition, the player is encouraged to merely
play around with these features. Once the player gets bored with a
certain realm or epoch, it's no more satisfying to forcibly move the
PCs than to simply quit and restart (which is a moot point, since
movement between realms hasn't been implemented, either).
This would all be irrelevant to the discussion, except that the
question arises: could (or should) a roguelike game be structured this
way? Certainly, not *every* one, or even *any* one, since the
necessary prerequisite is the motivation of the developer. However,
it's been shown throughout this thread that some subset of gamer is
dissatisfied with several features of many roguelike games.
All current roguelike games have their features structured vertically.
ADOM has its classes gain powers at levels throughout the game.
Nethack ends with the PC going through hell, the elemental planes, and
heaven, but starts with a bland dungeon. I could go on, but the
picture is fairly obvious. Would it be interesting to play a roguelike
game that's structured horizontally? I certainly think it would be
good, if only just for a change of pace, since dungeon crawls and
powergaming still have a large place in the roguelike gamer community.
Obviously, the audience for such a game is much different from that of
'normal' roguelike games. However, I think it's worth a discussion,
since it involves paradigms beyond the "XP or Skills" controversy
that's typical of this group.
Gerry Quinn
>David Damerell <dame...@chiark.greenend.org.uk> wrote in message
> news:<VEv*ai...@news.chiark.greenend.org.uk>...
>>
>> What I meant was that the OP appeared to be making some design decisions
>> purely in order to prevent save-scumming; decisions that might even have
>> other negative consequences. I think that's a Bad Thing.
>
>As far as I'm concerned, save-scumming is just a symptom of the
>problem. If a player is writing scripts or copying files around as
>part of playing the game, it's showing a high level of
>dissatisfaction.
It can become addictive. You start by saying "I'll save my progress
now, I have gone so far". Then "Well, I picked up some nice goodies in
the last level, I think I'll save." But the end result is a pathetic
save-reload junkie who saves before opening every chest. Not everyone
who saves-reloads gets to this stage, but many do.
Drug addiction shows a high degree of dissatisfaction with life, but it
feeds on its own dissatisfaction - it's not life that's the problem.
>Besides, you don't have to be such an idiot. Mainstream games have
>difficulty levels. No roguelike game has them. What does that say?
>Roguelike developers have been ignoring (and continue to ignore)
>possible features that could improve their games.
Many mainstream CRPGs don't have difficulty levels. There is a
compensating issue, particular in roguelikes, because YOU choose how
deep you go. In a sense, you are ALWAYS selecting difficulty level in a
typical roguelike - every time you choose between an up or a down
staircase you are adjusting the difficulty.
Gerry Quinn
--
http://bindweed.com
Entertainment software for Windows
Games, Puzzles, Kaleidoscope Screensavers
Download evaluation versions free - no time limits
Martin Read
Amy Wang <blueme...@hotmail.com> wrote:
>David Damerell <dame...@chiark.greenend.org.uk> wrote in message news:<VEv*ai...@news.chiark.greenend.org.uk>...
>> What I meant was that the OP appeared to be making some design decisions
>> purely in order to prevent save-scumming; decisions that might even have
>> other negative consequences. I think that's a Bad Thing.
>
>As far as I'm concerned, save-scumming is just a symptom of the
>problem. If a player is writing scripts or copying files around as
>part of playing the game, it's showing a high level of
>dissatisfaction.
Or possibly of incompetence.
>Besides, you don't have to be such an idiot. Mainstream games have
>difficulty levels.
*Some* do. As an example from the wargames arena, most of the Heroes III
campaigns don't. In the CRPG environment - well, Crusaders of the Dark
Savant had them, but the best way to make the game easy was to play on
normal, rather than easy or expert (fighting certain setpiece encounters
on Expert was officially Not Fun.), all the time. (if you were minded to
cheat slightly, you played on Expert most of the time and dropped to Easy
for the setpiece encounters :)
>No roguelike game has them. What does that say?
>Roguelike developers have been ignoring (and continue to ignore)
>possible features that could improve their games.
Roguelike developers are, by and large, not ignoring them. They are
taking note of them, and deciding that they don't want them.
Ignore != consider and discard.
>If you're just going to dismiss a widely accepted feature as being a
>'Bad Thing' due to some vague allusions to negative consequences, then
>that's your arguement. I've put mine down already.
Here's *my* experience of the savescumming thing.
I *used* to savescum (and do other evil things like wizard-mode bones
piles) in Nethack (and in Hack; somehow I resisted doing so in rogue.)
I didn't become any *good* at Nethack until I stopped.
m.
--
\_\/_/| Martin Read - my opinions are my own. share them if you wish.
\ / | the conference hall rings to the standing ovation the people in blue
\/ | ties rise from the podium crazy with power blinded by vision the mass-
------+ chosen leaders for a brutalised nation -- NMA, "Archway Towers"
Eytan Zweig
> >difficulty levels. No roguelike game has them. What does that say?
That you're not familiar with enough Roguelike games to know that that's a
false statement?Both EyAngband and Oangband, for instance, have difficulty
levels (though Oangband's ones are cleverly disguised as races, they clearly
are labelled as difficulty levels on the race selection screen and in the
manual); and so do other Angband variants. I'm not familiar with non-Angband
RLs much these days, but I'd be surprised if your huge generality is true
for all of them.
Eytan
David Damerell
>>Amy Wang:
>>>Besides, you don't have to be such an idiot. Mainstream games have
>>>difficulty levels. No roguelike game has them. What does that say?
>are labelled as difficulty levels on the race selection screen and in the
>manual); and so do other Angband variants. I'm not familiar with non-Angband
>RLs much these days, but I'd be surprised if your huge generality is true
>for all of them.
NetHack's classes provide difficulty levels, and it did in the past have
an "easy" setting.
Michael Blackney
>
> pau...@mbnet.fi (Paul Pekkarinen) wrote in message
news:<8f2c2bbc.0301...@posting.google.com>...
> > blueme...@hotmail.com (Amy Wang) wrote in message
> > > These all seem like good ways of increasing player enjoyment. I
> > > wouldn't have thought of giving the PC better equipment. Although it
> > > is a one-time benefit, the potential of giving the player a decent
> > > chance at experiencing the game is attractive.
> >
> > Well this better equipment has already invented (with some other
> > benefits) in so called player profession or class. It's
> > easier to play Nethack with barbarian than tourist.
>
> Didn't I already explain why you're wrong? I guess I'll try again. The
> barbarian is not equivalent to a wizard with lower difficulty. They
> are different classes, implying a different way of playing the game.
> The fact that the barbarian is 'easier' is merely due to the fact that
> Nethack's developers spent no time on play balance, since the entire
> point of Nethack is to have an unfair challenge. Why don't you stop
> trying to dodge my point and argue against it if you don't like it?
>
He isn't wrong at all. If I am a wrestler and am getting used to fighting
against people I can beat, I have several options to challenge myself.
1. Wrestle against someone stronger.
2. Wrestle against two people at once.
3. Wrestle with the same opponents, but with one arm tied behind my back.
In general it's much more impressive to do the third, and it is this type
of difficulty system that most roguelikes implement. And in Nethack,
besides the differing difficulties that classes impose, there are the
challenge score cards which help to add even more challenge and make the
game even more interesting.
Oh, but I guess you were only wanting to make the games *easier*.
> > Who says we need difficulty levels like mentioned earlier?
>
> I do. Several people here do. Every gamer in the entire world, with
> the exception of a few 'hardcore' roguelike gamers, does. Like I said
> before, however, I'm not saying that every game needs to have
> difficulty levels. I'm merely saying that there are *many* cases in
> which it would improve a game to have this feature.
>
I'm not hardcore and I don't agree with your difficulty level idea.
Actually, I think it's quite a stupid idea for a roguelike.
> > And why roguelikes (or other games) should be made thinking those
> > wimps who cheat all the time in all games? "Pat pat, we'll make this
> > easy for you.. don't worry."
>
> That's not even a sentence. I'm not going to bother trying to
> extrapolate some kind of meaning from it, either, since it is
> undoubtedly just an adolescent attempt at diffusing an argument with
> childish whining.
That's not even an argument. I'm not going to bother trying to
extrapolate a valid and relevant argument from it, either, since it is
undoubtedly just an adolescent attempt at diffusing an argument with
childish whining.
--
michaelblackney at hotmail dot com
Worst roguelike ever: http://www27.brinkster.com/atrl/
R Dan Henry
<ez...@nyu.edu> declared:
>"DarkGod" <dar...@ifrance.com> wrote in message
>news:pan.2003.01.20....@ifrance.com...
>> While under the effect of mushrooms of hallucination "someog"
>> <som...@mailcity.com> wrote:
[scumming Angband]
>>> Huh? I though the number of HP you got was a non-random function of
>>> your CON and cLevel. Am I wrong?
>> It doth have a random factor
>But it's a pre-determined random factor; it's impossible to alter it by
>save-scumming.
This is true now, but long ago, so long ago that I think it was
actually Moria rather than Angband, it was not. Hit points were not
rolled in advance and it was possible to lose a level, regain the
level, and the hit points would be rerolled for the regained level.
This did not require cheating, merely abuse. Grape jellies were the
preferred method of scumming higher hit points, as they could drain
experience but were immobile and therefore almost harmless.
Note that this was an ancient bug, long fixed. It is possible that it
was in very early Angband, but I'm fairly certain that it was not an
issue at least by the "frog-knows" releases.
--
R. Dan Henry
rdan...@earthlink.net
They can have my ASCII graphics when they pry them
from my cold dead (c) and (d) slots.
R Dan Henry
<gr...@puyo.cjb.net> declared:
>Mine is not a traditional roguelike! I know it makes you squirm and hate
>me that I'm writing something that may more adequately be described as a
>randomised, text-based, party-based, plot-heavy, Nintendo and old school
>Squaresoft inspired CRPG. *G* It's a different *type* of game to your
>average roguelike, so talking about balance assumes a whole lot which
>doesn't really translate.
Doesn't make me hate you. Just makes me uninterested. (Uninterested in
playing the game, that is. Discussion of specific design issues may
still be very interesting to me.)
R Dan Henry
<gr...@puyo.cjb.net> declared:
>Do you find the save/load system in Zelda games annoying? I don't.
>That's because a vast majority of the time, I'm happy to play the game
>until I reach a safe point to save, where I will be returned when I
>restart the game. It's only when I urgently need to leave the game that
>I'll save and exit knowing that I'll have to retrace my steps to get
>back to where I currently am (e.g. I'm running late for uni, or my
>sibling wants to use the computer for work). These are exceptional
>circumstances, at least in my life.
Needing to go do something other than play a computer game and not
being able to get back to it soon is not exceptional in the lives of
many others, whom you might want as players. I can't say I'd think
much of a game that doesn't let me save where I am right now.
I also fairly frequently stop playing because I'm getting tired or
being distracted, neither of which are good for my playing skills.
This does not happen on schedule.
Greg McIntyre
> Many mainstream CRPGs don't have difficulty levels. There is a
> compensating issue, particular in roguelikes, because YOU choose how
> deep you go. In a sense, you are ALWAYS selecting difficulty level in
> a typical roguelike - every time you choose between an up or a down
> staircase you are adjusting the difficulty.
Tempted to scream "WHAT!?" but I'll just stick with a sound 'no'.
Mark 'Kamikaze' Hughes
> ger...@indigo.ie (Gerry Quinn) wrote:
>> Many mainstream CRPGs don't have difficulty levels. There is a
>> compensating issue, particular in roguelikes, because YOU choose how
>> deep you go. In a sense, you are ALWAYS selecting difficulty level in
>> a typical roguelike - every time you choose between an up or a down
>> staircase you are adjusting the difficulty.
> Tempted to scream "WHAT!?" but I'll just stick with a sound 'no'.
No, wait, it's true. You have the choice of dying of starvation on an
emptied level or finding new stuff but risking death. Dying of
starvation is easy. |+)
--
<a href="http://kuoi.asui.uidaho.edu/~kamikaze/"> Mark Hughes </a>
"We remain convinced that this is the best defensive posture to adopt in
order to minimize casualties when the Great Old Ones return from beyond
the stars to eat our brains." -Charlie Stross, _The Concrete Jungle_
Gerry Quinn
>Tue, 28 Jan 2003 19:07:19 +1100, Greg McIntyre <gr...@puyo.cjb.net>:
>> ger...@indigo.ie (Gerry Quinn) wrote:
>>> Many mainstream CRPGs don't have difficulty levels. There is a
>>> compensating issue, particular in roguelikes, because YOU choose how
>>> deep you go. In a sense, you are ALWAYS selecting difficulty level in
>>> a typical roguelike - every time you choose between an up or a down
>>> staircase you are adjusting the difficulty.
>> Tempted to scream "WHAT!?" but I'll just stick with a sound 'no'.
>
> No, wait, it's true. You have the choice of dying of starvation on an
>emptied level or finding new stuff but risking death. Dying of
>starvation is easy. |+)
You are right as regards some roguelikes such as NetHack or
(particularly) Rogue, but in others such as Angband and Moria, what I
said is true. You can hang around in level X of Angband forever - just
go upstairs and down again, and the level is completely replenished.
Greg McIntyre
> You are right as regards some roguelikes such as NetHack or
> (particularly) Rogue, but in others such as Angband and Moria, what I
> said is true. You can hang around in level X of Angband forever -
> just go upstairs and down again, and the level is completely
> replenished.
Erm... If you ask me, true but pointless. I wouldn't find playing
through the same level over and over again to be enjoyable in the
slightest.
Gerry Quinn
>ger...@indigo.ie (Gerry Quinn) wrote:
>> You are right as regards some roguelikes such as NetHack or
>> (particularly) Rogue, but in others such as Angband and Moria, what I
>> said is true. You can hang around in level X of Angband forever -
>> just go upstairs and down again, and the level is completely
>> replenished.
>
>Erm... If you ask me, true but pointless. I wouldn't find playing
>through the same level over and over again to be enjoyable in the
>slightest.
By replenished, I meant a new level is created, with a new map, monsters
and items. In other words, you can stay on Level X difficulty as long
as you want.
be...@sonic.net
>
> ger...@indigo.ie (Gerry Quinn) wrote:
> > Many mainstream CRPGs don't have difficulty levels. There is a
> > compensating issue, particular in roguelikes, because YOU choose how
> > deep you go. In a sense, you are ALWAYS selecting difficulty level in
> > a typical roguelike - every time you choose between an up or a down
> > staircase you are adjusting the difficulty.
>
> Tempted to scream "WHAT!?" but I'll just stick with a sound 'no'.
>
Hmmm. You could make the "easy" game the game where you just stop.
Put a somewhat-badder-than-average monster on level 5 and have the
dungeon be only five levels deep. The game concludes, the player
gets closure, and the player doesn't get a peek at what lurks on
level 80.
Bear
Greg McIntyre
> By replenished, I meant a new level is created, with a new map,
> monsters and items. In other words, you can stay on Level X
> difficulty as long as you want.
Until you get so utterly bored/frustrated because you're not making any
progress at all that you quit the game and never play it again.
If you're not working towards the end of the game, you're just dicking
about in it. I don't understand what the point of that is. Some games
are enjoyable to just *play* (ala Pong), but I don't find roguelikes to
be so. Roguelikes are fun for me because there's a constant progression
-- your character is constantly improving, and you have the lure of
fantastic treasure and heroic battles ahead of you. If you stay on level
X, your character doesn't improve much, you have the prospect of
mediochre treasure, and battles of the calibre you've already won.
Greg McIntyre
> Hmmm. You could make the "easy" game the game where you just stop.
> Put a somewhat-badder-than-average monster on level 5 and have the
> dungeon be only five levels deep. The game concludes, the player
> gets closure, and the player doesn't get a peek at what lurks on
> level 80.
That would be an improvement, in that new players who tend to die on
level 6 would get some closure and satisfaction at level 5, and so
wouldn't get totally disheartened. :)
Gerry Quinn
Tbe point is, you can stay there UNTIL you feel ready to tackle level
X+1. All players of Angband / Moria decide how long to stay on
intermediate levels - always going down the first down staircase is a
recipe for rapid death.
You are working towards the end anyway - you are collecting useful wands
and potions - and the knowledge of which is which. (IIRC, potions of
speed are actually most common on level 2 of Zangband.)
R Dan Henry
<gr...@puyo.cjb.net> declared:
>ger...@indigo.ie (Gerry Quinn) wrote:
>> You are right as regards some roguelikes such as NetHack or
>> (particularly) Rogue, but in others such as Angband and Moria, what I
>> said is true. You can hang around in level X of Angband forever -
>> just go upstairs and down again, and the level is completely
>> replenished.
>
>Erm... If you ask me, true but pointless. I wouldn't find playing
>through the same level over and over again to be enjoyable in the
>slightest.
Neither would I. I will dive when the current level is no longer
interesting to be playing at, but in between the "take the first down
stairs" ultraIronmen and the Borg, lie a variety of players, some of
whom seem quite happy to sweep over the same depth again and again.
Amy Wang
> "Amy Wang" <blueme...@hotmail.com> communicated:
> >
> > Didn't I already explain why you're wrong? I guess I'll try again. The
> > barbarian is not equivalent to a wizard with lower difficulty. They
> > are different classes, implying a different way of playing the game.
> > The fact that the barbarian is 'easier' is merely due to the fact that
> > Nethack's developers spent no time on play balance, since the entire
> > point of Nethack is to have an unfair challenge. Why don't you stop
> > trying to dodge my point and argue against it if you don't like it?
> >
>
> He isn't wrong at all. If I am a wrestler and am getting used to fighting
> against people I can beat, I have several options to challenge myself.
>
> 1. Wrestle against someone stronger.
> 2. Wrestle against two people at once.
> 3. Wrestle with the same opponents, but with one arm tied behind my back.
>
> In general it's much more impressive to do the third, and it is this type
> of difficulty system that most roguelikes implement. And in Nethack,
> besides the differing difficulties that classes impose, there are the
> challenge score cards which help to add even more challenge and make the
> game even more interesting.
>
> Oh, but I guess you were only wanting to make the games *easier*.
I want a player to be able to consciously select a difficulty level
without having to learn how the game balance is skewed. If a
commercial game were released with such a lazy "feature" that exists
only for apologetics like you to diffuse pro-difficulty level
arguments, the game would be slammed by the critics.
While option #3 is good (although the analogy is flawed), is there
anything inherently wrong with #1 or #2? Continuing with Nethack,
what's wrong with a player who wants to play as a wizard more easily,
or as a barbarian with more difficulty? These extra challenges *ARE
NOT* the same as difficulty levels. I believe that people here
understand that, and if you insist that they are, you'll just end up
looking like a fool.
>
> I'm not hardcore and I don't agree with your difficulty level idea.
> Actually, I think it's quite a stupid idea for a roguelike.
>
WHY? Why don't you just give the group a valid reason why this idea is
bad, instead of saying it's already been done?
R Dan Henry
<gr...@puyo.cjb.net> declared:
>be...@sonic.net wrote:
>> Hmmm. You could make the "easy" game the game where you just stop.
>> Put a somewhat-badder-than-average monster on level 5 and have the
>> dungeon be only five levels deep. The game concludes, the player
>> gets closure, and the player doesn't get a peek at what lurks on
>> level 80.
>
>That would be an improvement, in that new players who tend to die on
>level 6 would get some closure and satisfaction at level 5, and so
>wouldn't get totally disheartened. :)
Some Angband players have reported a satisfying closure in killing a
unique for the first time after losing several previous characters to
that unique. Providing intermediate success marks between the start
and end of the game may be helpful for some players to feel
satisfaction in a non-winning game. I certainly consider my current
Crawl game to be a "success" in spite of being nowhere near
completion. In addition to reaching higher levels, character and
dungeon, than ever before, this character has also penetrated dungeon
branches I'd not reached before. As well as taking revenge on a couple
of uniques that have been trouble in the past.
R Dan Henry
<ez...@nyu.edu> declared:
>"someog" <som...@mailcity.com> wrote in message
>news:b096u3$mk9tt$1...@ID-175339.news.dfncis.de...
>> "Amy Wang" <blueme...@hotmail.com> wrote in message
>> news:62acd54d.03011...@posting.google.com...
>> > On the other hand, would it have been worth it to spend the same
>> > number of hours and *not* win? Angband encourages save-scumming, even
>> > if its developer didn't intend it to. You're given no real reason to
>> > play after winning, so the game has replay value in its ability to
>> > prevent players from winning. There's no plot to appreciate, or even
>> > the special features found in other roguelikes.
If I was enjoying playing the game, then *of course* it would have
been worth it. If I wasn't enjoying it, then playing on to win isn't
going to be much fun anyway. The difference between play and work is
that play is valued for the *process* and not simply the *result*.
>That's true of some players, but hardly of all players. As I said in an
>earlier message, I find plot to be a very strong encouragement to savescum -
>because, unless plot is done really, really, well, I don't get immersed
>enough to feel part of it; for 99% of games with good plots, I think "I'm
>being told an interesting story" and then I try to get as much of the story
>as possible, rather than play by the rules of the story. Savescumming is a
>tool to see more of the plot.
I agree that plotted games do more to encourage savefile abuse. After
all, it encourages one to see the game as a story and in a story one
expects to move on to the conclusion. A plot isn't simply a set of
plot elements. A plot is a *structure* of events. As such the "random
plot" concept can only deserve that name if it is filling in parts of
a script (in the literary, rather than the programming sense of the
word). I'm aiming for something looser and think others may be, too, a
"complex simulation" or "living world" that allows for events to build
on one another and for NPCs whose interests collide and interact in
ways that create interesting situations. But not a true plot. Even
something as loose a ADOM's quest structure, where one doesn't
strictly have to do the quests, adds a "been there, done that" feeling
to restarting the game. With a traditional plotted game, there's a
huge incentive to restore rather than accept death. You've already
been through that part of the story. If the replaying the first part
of the game is like rereading the first two chapter of a novel, it
becomes very unappealing.
>Basically, I think that any game in which Amy will be encouraged to savescum
>I won't be, and vice-versa. And there are probably hundrends of different
>attitudes that other players may have. That's the real challange on people
>who seek to prevent save-scumming by gameplay principles.
Well, I think a certain flexibility can help minimize the temptation.
Give the player more options about gameplay (Do I have to kill the
evil minion and the evil overlord or can I just find the legendary
hero and let him do the dirty work? Can I select from a set of
challenges I want to try, like vegetarianism or weaponless? Are there
easier and harder races and classes to choose from? Can I play either
freeform or follow a quest with a plot structure?) and there's less
reason to try to circumvent the game as designed. OTOH, the more you
try to broaden things, the more difficult it is to balance things.
Gameplay options always represent problems as well as solutions.
>> Yes, it would be worth it, if the reason I play (and it is) is to measure
>> my skills against the difficulty of the game.
That's one reason. Others might be that I like collecting slime molds
(personally, I don't, I think it's a strange hobby, but others have
been quite amused by their slime molds) and that doesn't require
getting that far. I might enjoy trying lots of different character
types and not at all mind starting a new one. I might just like the
individual tactical contests and not get much out of the long-term
goal structure.
>From here onwards, though I no longer agree with David's reasoning:
>> Consider these ideas, and I doubt you'll disagree:
>>
>> 1) Save-scummers are either unable or unwilling to beat the game without
>> save-scumming.
>> 2) Save-scummers who beat the game by so doing do not deserve the
>accolades
>> that come with winning.
>> 3) Save-scummers who beat the game by so doing (for whatever reason), and
>> who do not ask for these accolades, are harmless. We should even
>accomodate
>> them if possible in their innocent diversions, since they harm no one.
>> 4) Save-scumming protection schemes are generally built to thwart the
>> efforts of those in point 2.
>> 5) There exists a mechanism by which those in point 3 can accomplish their
>> goal without save-scumming. In fact, they can do so with even less effort
>> than save-scumming requires. It's called cheating death, and it's already
>> available. The only thing they give up is what they didn't want anyway;
>the
>> ability to post their (invalid) scores.
>>
>> So why fight those who implement save-scumming restrictions?
>>
>
>You seem to be underestimating the power of save-scumming. Save-scumming
>isn't just there to cheat death. Back when I used to save-scum, I scummed to
>get better drops, I scummed to avoid minor effects such as stat drain, I
>ID'd items by experimentation and scummed if I ended up causing bad effects,
>and sometimes ID'd items by selling them and scummed if I got a bad deal.
>Now, I was taking it to an extreme that most save-scummers probably don't,
>but that doesn't mean some of them don't do part of that.
This is why "restore after death" should not be called savescumming.
It is a cheat, but it isn't a form of scumming. True savescumming is
cheesier beyond the mere cheat-death level.
And it isn't just in roguelikes. I've read "tips" for Civilization
that consisted of nothing but extremely cheesy savescumming ("if you
don't get a civilization advance or free city when you explore a
village, restore and try again"). And because non-roguelikes don't try
to stop it, it doesn't even occur to some of these players that their
"clever strategies" are just cheating their way around the game.
>Note that I'm not arguing any particular point here, just pointing out that
>the situation is more complex than presented. The entire issue of preventing
>save-scumming seems to me fine in principle, but in practice it's a waste of
>time and energy.
Well, I certainly agree so far as *enforcing* savefile abuse goes, but
I don't think efforts to prevent it are wasted. A good, clear
explanation of why you actually die when killed in the game ought to
be in the basic game manual. Instead of trying to prevent a
replacement savefile from being used, why not just detect it and issue
a warning that doing so for reasons other than a game crash/power
failure is cheating? If someone wants to cheat, you aren't going to
stop him, but you can try to make him actually think about and decide
if he really wants to.
Joseph Hewitt
> He'd scum until he got the maximum hit point rolls when
> he went up a level - every level.
> ...
> He'd even dig in the rubble, not find anything, and scum to dig
> in the rubble again until he found something good.
It seems to me that this type of save-scumming may have something to
do with the reason why people get addicted to slot machines... simple
repeditive behavior, an irregular reward schedule, yadda yadda
yadda... If I had my psych textbooks here I might be able to write up
an explination that sounds halfway plausible.
It also seems to me that some of the random outcome features in common
RLs- magic pools in ADOM, scrolls of acquirement, that kind of thing-
may well promote this kind of behavior in those players who are
succeptable to it.
- Joseph Hewitt
--
DeadCold > http://www.geocities.com/pyrrho12/programming/deadcold/index.html
GearHead > http://www.geocities.com/pyrrho12/programming/gearhead/index.html