Spelunky - a platformer roguelike?
Nathaniel Rook
game; I just find it fascinating.
Anyway, Spelunky is a new game made by someone who, as far as I know,
has never even heard of rec.games.roguelike.development, and, indeed,
has probably not played many roguelikes. As such, it very much has an
"old platform game" aesthetic, but it's still undeniably a roguelike,
with conservation of resources, tactical offensives, and a fear of
death playing predominant roles. Unfortunately, it's closed source
and Windows only, but I figure it's still worth playing if possible.
http://forums.tigsource.com/index.php?topic=4017.0
-Nathaniel Rook
HED
culture". Here's his wikipedia entry : http://en.wikipedia.org/wiki/Derek_Yu
Jeff Lait
>
> Anyway, Spelunky is a new game made by someone who, as far as I know,
> has never even heard of rec.games.roguelike.development, and, indeed,
> has probably not played many roguelikes.
I don't understand these disclaimers. It makes it sound like this
game shares roguelike elements due to chance convergent evolution.
But, the author in the page you link to says:
"My goal was to create a fast-paced platform game that had the kind of
tension, re-playability, and variety of a roguelike."
It seems clear that this game is explicitly inspired from roguelikes,
so considering it at a minimum a roguelike-like is a no-brainer.
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)
Nathaniel Rook
I mean merely that Yu is not a roguelike developer by trade, and,
thus, Spelunky takes a lot more from outside the "sphere of roguelike
development" than most new roguelikes do.
Really, I find it interesting because it's very obviously a roguelike,
but it doesn't have a lot of the common roguelike features - thus, it
illuminates some things which are disposable in a roguelike, but are
generally taken for granted.
Well, that's not true. I find it interesting because I can't stop
playing it!
-Nathaniel Rook
Dark_Oppressor
(ascii or tile graphics, turn based, etc.) While I love such games,
it seems to be that many people are not very interested. I know that
I can get more friends to try something like Spelunky than Nethack.
While this is frustrating, it is still a fact. Perhaps if more people
took some of the (in my humble opinion) best features of roguelikes
and used them in other types of games, more people could be introduced
to the roguelike genre. Things like randomly generated levels,
enemies, items, etc. are great features for many games, not just
roguelikes. One of the aspects of roguelike games that I have run
into the most hatred for among my friends is perma death.
Perhaps a game that takes many, or even most, of the best aspects of
roguelikes but excludes perma death could reach popularity on a larger
scale. I personally love permanent deaths, but it seems that many
people outside of the roguelike "world" don't. I suppose I am just
rambling, but I love roguelikes, and the prospect of games that might
reach out to more people and show them how awesome roguelikes can be
is quite exciting.
pau...@mbnet.fi
> Really, I find it interesting because it's very obviously a roguelike,
It's not a roguelike. It's a platform game with random content.
I think some roguelike developers are like hammers: they see
only nails. So when there is a game with some feature known
from roguelikes (ascii, randomness, permadeath, etc.) they
make it a roguelike.
Slash
> game; I just find it fascinating.
>
> Anyway, Spelunky is a new game made by someone who, as far as I know,
> has never even heard of rec.games.roguelike.development, and, indeed,
> has probably not played many roguelikes.
Mr. Derek Yu has knowledge of all the indie scene! that includes rgrd
and roguetemple. He was a great knowledge about roguelikes.
He also *seems to* have posted here back on 2001 (search and you will
find)
But, I agree, he is NOT a roguelike developer at all. Actually, he is
not a raw developer like most of us but he's also a great artist (Ok,
he is actually more of an artist than a programmer, see the credits
for his masterpiece, Aquaria)
> As such, it very much has an
> "old platform game" aesthetic, but it's still undeniably a roguelike,
> with conservation of resources, tactical offensives, and a fear of
> death playing predominant roles. Unfortunately, it's closed source
> and Windows only, but I figure it's still worth playing if possible.
I noticed about it, but have been a bit too busy to play it, if anyone
wants to review it for roguetemple, please contact me!
Jeff Lait
>
> Perhaps a game that takes many, or even most, of the best aspects of
> roguelikes but excludes perma death could reach popularity on a larger
> scale.
No need for "Perhaps". We already know the answer to this. I
understand the third iteration of the Diablo franchise is coming soon
now too.
OTOH, permanent death, meaning "start over from level 1", is what
drives most of the other roguelike characteristics.
Procedural level generation is a *bad* idea if you never reset. If
the player is expected to play through once, you can get much better
results by hand crafting.
Diablo is not quite the counter example people may think it is. When
you play a level you don't save part way. You always reset the level
you are working on.
> I personally love permanent deaths, but it seems that many
> people outside of the roguelike "world" don't. I suppose I am just
> rambling, but I love roguelikes, and the prospect of games that might
> reach out to more people and show them how awesome roguelikes can be
> is quite exciting.
I think it would be a more interesting challenge to take the permanent
death aspect of roguelikes and move it into other genres. An FPS
version of the roguelike Urban Warfare would be better than many of
the "memorize sniper location" FPS escapes we are subjected to.
People hate permanent death in games because it was often used by lazy
designers as a way to force you to spend more hours playing the game.
There was no virtue to it, it added nothing to the game. A roguelike
is different in that it is more enjoyable with permanent death than
without. Trying to play nethack in explore mode is an exercise in
frustration to a newbie - the game seems utterly impossible.
The key rule for a permanent death game is to make level 1 always
interesting. This is *why* roguelikes go for procedural generation,
unidentified items, etc.
Gerry Quinn
@r15g2000prh.googlegroups.com>, torespon...@hotmail.com says...
> Diablo is not quite the counter example people may think it is. When
> you play a level you don't save part way. You always reset the level
> you are working on.
In fact all levels are reset; you can go back to a previous level and it
will be a randomised version similar to the first time you were there
(similar to Angband). The only difference is that quests can only be
completed once.
Diablo also had the 'ironman' option of perma-death. Some people liked
it, or claimed to...
- Gerry Quinn
--
Lair of the Demon Ape (a coffee-break roguelike)
<http://indigo.ie/~gerryq/lair/lair.htm>
Jotaf
> Diablo also had the 'ironman' option of perma-death. Some people liked
> it, or claimed to...
For a long time I thought that the solution was to make perma-death
kinda optional. But when I played Diablo (and maybe other games with
this option) I never even considered that mode. I thought it was for
crazy people who had played the game for a long time. I only saw it as
a fair game mode when I stumbled upon the RL genre.
I understand that resetting is important, but some features such as
Ankhs to resurrect the player are a great bonus for advanced
characters. If it was simply a fixed game point at level 20 instead of
a randomized bonus the player's strategy would change dramatically,
like when playing a game with save points such as Final Fantasy.
Jotaf
Andrew Doull
I think what makes it different from roguelikes is intriguing.
Firstly, the maps have a 'grain' to them, caused by gravity, which the
player is able to reverse a limited number of times (using a rope) but
ultimately implies a direction of travel (and which is why all map
exits are 0, 1 or 2 grids from the ground, that being the maximum
height the player can get unaided).
Secondly, while the game is ostensibly real time, it is able to get
away with the convention that almost all enemy encounters are
initiated by the player that most platformers have: you rarely have to
fight an enemy you don't make the decision to. So the real time-iness
is actually limited to very small chunks of play, most of which
consist of timing jumps or attacks.
Andrew
goblinhacker
randomly? I'm wondering how you guarantee to make a level solvable
I'm about to add the same thing to goblinhack - I think what he has is
really nice
tx
Neil
David Ploog
[snip explanation how permadeath cripples success of roguelikes most,
citing Diable as a successful roguelike game without it]
> OTOH, permanent death, meaning "start over from level 1", is what
> drives most of the other roguelike characteristics.
This is true. In my opinion, making a (classical) roguelike is about how
to make starting again interesting, ie. to eke out most of permadeath.
> Procedural level generation is a *bad* idea if you never reset. If
> the player is expected to play through once, you can get much better
> results by hand crafting.
Yes. One can try to combine both approaches. Crawl draws from a large
number of random vaults: no matter if you go for theme and/or challenge,
hand made will provide better results. Once your pool is large enough, you
get the same old randomness.
> People hate permanent death in games because it was often used by lazy
> designers as a way to force you to spend more hours playing the game.
> There was no virtue to it, it added nothing to the game.
Having spend hours at various classical scrolling shooters (Darius Gaiden
and their ilk), I beg to disagree. By replaying the earlier parts of the
game again and again, you become better, thus arriving at the challenging
parts with more powerups etc. In other words, additional motivation is
gained by additional challenges: not just "get to level 3" but "get to
level 3 with all shields collected" etc.
> The key rule for a permanent death game is to make level 1 always
> interesting. This is *why* roguelikes go for procedural generation,
> unidentified items, etc.
That nicely sums it up. Actually, in a (vain?) attempt to make level 1
more interesting, Crawl uses a huge number of entry vaults (about 150 or
so). Some players do not like it, but I think it helps more than it hurts.
David
konijn_
One thing this game made me realize is that a game without means to
restore hitpoints is not a roguelike for me. And not fun enough to
play for years in a row.
T.
paladi...@gmail.com
There is a way to restore health. When you rescue the gals, who as
the stereotypical blond went in and got lost) they will give you a
kiss, increasing your health by one. Also, there is a shop that will
allow you to gain another hp if you are willing to pay.
Brog
> On 7 tammi, 21:55, Nathaniel Rook <medi....@gmail.com> wrote:
>
> > Really, I find it interesting because it's very obviously a roguelike,
>
> It's not a roguelike. It's a platform game with random content.
And permadeath, gold collection, resource management, inexplicable
shops in the dungeon, learning through repeated failure, items with
multiple non-obvious uses, unidentified items, undocumented secrets,
unusual strategies which players use to mitigate the damage of
possible random effects (e.g. whipping pots from a distance instead of
throwing them; I liken this to the way ADOM players always kick doors
from the diagonal), and a general "exploring underground caves" feel.
It's not just roguelike developers who find that this game has a very
roguelike feel to it.
On Jan 9, 8:08 am, Andrew Doull <andrewdo...@gmail.com> wrote:
>
> Secondly, while the game is ostensibly real time, it is able to get
> away with the convention that almost all enemy encounters are
> initiated by the player that most platformers have: you rarely have to
> fight an enemy you don't make the decision to. So the real time-iness
> is actually limited to very small chunks of play, most of which
> consist of timing jumps or attacks.
It actually has another real time element, which chances are you never
came across - if you spend more than about two minutes on a level a
ghost will show up which kills you instantly if you touch it. This is
analogous to Rogue's food timer - you can't spend as long as you want
on a level collecting every bit of gold, you have to keep on moving
deeper in. It's not a big part of the game because it's a very
generous time limit, you have to muck around wasting a lot of time or
juggling several items to meet up with it, but a lot of noobs complain
about it still.
Curiously, I haven't seen many complaints about the permadeath.
On Jan 9, 12:32 pm, goblinhacker <goblinh...@gmail.com> wrote:
> Does anyone know the algorithm he uses to make the platform levels
> randomly? I'm wondering how you guarantee to make a level solvable
I believe it's largely based on having a large number of set pieces
which are combined randomly, with maybe a bit of variation within.
I'm not sure what method he uses to guarantee solubility, but one
simple method is just to search for a valid path from start to end.
The fact that you're given ropes and bombs helps a bit - even if there
isn't a path you can make one.
Curiously, Spelunky does not always generate solvable levels (apart
from use of ropes/bombs/mining), especially in the later areas.
konijn_
> Does anyone know the algorithm he uses to make the platform levels
> randomly? I'm wondering how you guarantee to make a level solvable
http://www.squidi.net/three/entry.php?id=84
Obviously, that's just a start.
T.
<snip>