The Berlin Interpretation

384 views
Skip to first unread message

Jeff Lait

unread,
Sep 26, 2008, 4:13:09 PM9/26/08
to
==Preamble==

This definition of "Roguelike" was created at the International
Roguelike Development Conference 2008 and is the product of a
discussion between all who attended. The definition at
http://www.roguetemple.com/roguelike-definition/ was used as the
starting point for the discussions. Most factors are newly phrased,
new factors have been added, some factors have been removed.

==General Principles==

"Roguelike" refers to a genre, not merely "like-Rogue". The genre is
represented by its canon. The canon for Roguelikes is ADOM, Angband,
Crawl, Nethack, and Rogue.

This list can be used to determine how roguelike a game is. Missing
some points does not mean the game is not a roguelike. Likewise,
possessing some points does not mean the game is a roguelike.

The purpose of the definition is for the roguelike community to better
understand what the community is studying. It is not to place
constraints on developers or games.

==High value factors==

====Random environment generation====

The game world is randomly generated in a way that increases
replayability. Appearance and placement of items is random.
Appearance of monsters is fixed, their placement is random.
Fixed content (plots or puzzles or vaults) removes randomness.

====Permadeath====

You are not expected to win the game with your first character. You
start over from the first level when you die. (It is possible to save
games but the savefile is deleted upon loading.) The random
environment makes this enjoyable rather than punishing.

====Turn-based====

Each command corresponds to a single action/movement. The game is not
sensitive to time, you can take your time to choose your action.

====Grid-based====

The world is represented by a uniform grid of tiles. Monsters (and
the player) take up one tile, regardless of size.

====Non-modal====

Movement, battle and other actions take place in the same mode. Every
action should be available at any point of the game. Violations to
this are ADOM's overworld or Angand's and Crawl's shops.

====Complexity====

The game has enough complexity to allow several solutions to common
goals. This is obtained by providing enough item/monster and item/item
interactions and is strongly connected to having just one mode.

====Resource management====

You have to manage your limited resources (e.g. food, healing potions)
and find uses for the resources you receive.

====Hack'n'slash====

Even though there can be much more to the game, killing lots of
monsters is a very important part of a roguelike. The game is player-
vs-world: there are no monster/monster relations (like enmities, or
diplomacy).

====Exploration and discovery====

The game requires careful exploration of the dungeon levels and
discovery of the usage of unidentified items. This has to be done anew
every time the player starts a new game.

==Low value factors==

====Single player character====

The player controls a single character. The game is player-centric,
the world is viewed through that one character and that character's
death is the end of the game.

====Monsters are similar to players====

Rules that apply to the player apply to monsters as well. They have
inventories, equipment, use items, cast spells etc.

====Tactical challenge====

You have to learn about the tactics before you can make any
significant progress. This process repeats itself, i.e. early game
knowledge is not enough to beat the late game. (Due to random
environments and permanent death, roguelikes are challenging to new
players.)

The game's focus is on providing tactical challenges (as opposed to
strategically working on the big picture, or solving puzzles).

====ASCII display====

The traditional display for roguelikes is to represent the tiled world
by ASCII characters.

====Dungeons====

Roguelikes contain dungeons, such as levels composed of rooms and
corridors.

====Numbers====

The numbers used to describe the character (hit points, attributes
etc.) are deliberately shown.

Darren Grey

unread,
Sep 27, 2008, 1:38:58 PM9/27/08
to
On Sep 26, 9:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:

> "Roguelike" refers to a genre, not merely "like-Rogue".  The genre is
> represented by its canon.  The canon for Roguelikes is ADOM, Angband,
> Crawl, Nethack, and Rogue.

So can we start a flamewar now? ;)

Considering you used those 5 games as the major canon I'm somewhat
surprised you haven't included other general features such as:

-Tolkien/D&D style fantasy setting. Obviously not necessary, but
extremely prevalent among roguelikes.

-Computer RPG gameplay elements. In particular stats such as HP,
strength, etc, and levels/experience gained through killing monsters.
Very few (if any) fully developed roguelikes don't have some sort of
character stat improvement system.

-Character creation that allows numerous very different starting
characters (usually through a race/class system). This ties in with
the random generation to increase replayability, since it allows very
different games each playthrough.

-Keyboard-only or keyboard dominant interface. This really ties in
with the ASCII display in that roguelikes have their roots in console
applications.

Perhaps some of these are too basic to mention though.

Of course personally I'm happy to call any game a roguelike if one of
the major influences on its creation was a roguelike or Rogue itself.
A member of the family by descent, as it were.

--
Darren Grey

Krice

unread,
Sep 27, 2008, 1:58:27 PM9/27/08
to
On 27 syys, 20:38, Darren Grey <darrenjohng...@gmail.com> wrote:
> -Character creation that allows numerous very different starting
> characters (usually through a race/class system).

This is actually the most important feature of a role-playing
game. They simply fucked it up, but don't worry, I'm going to
cure their delusions with my true definition.

dpeg

unread,
Sep 27, 2008, 8:20:55 PM9/27/08
to
Darren Grey wrote:

> On Sep 26, 9:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
>
>> "Roguelike" refers to a genre, not merely "like-Rogue".  The genre is
>> represented by its canon.  The canon for Roguelikes is ADOM, Angband,
>> Crawl, Nethack, and Rogue.
>
> So can we start a flamewar now? ;)

Of course! :)

> Considering you used those 5 games as the major canon I'm somewhat
> surprised you haven't included other general features such as:
>
> -Tolkien/D&D style fantasy setting. Obviously not necessary, but
> extremely prevalent among roguelikes.

This is nowhere needed for a roguelike. We tried to nail down properties
related to gameplay, and/or dealing with differentiating from other genres.
(And it might have been useful to state that.)
On the contrary (to your point), I'd rather say that the chosen
flavour/theme of a roguelike is arbitrary. You can easily transfer the
whole setting to something different without affecting the game as such.

> -Computer RPG gameplay elements. In particular stats such as HP,
> strength, etc, and levels/experience gained through killing monsters.
> Very few (if any) fully developed roguelikes don't have some sort of
> character stat improvement system.

We alluded to this in the last point: numbers. I think that's enough.
Again contrary (to your point), it might have been pointed out better that
roguelikes are _not_ roleplaying games, I think :)

> -Character creation that allows numerous very different starting
> characters (usually through a race/class system). This ties in with
> the random generation to increase replayability, since it allows very
> different games each playthrough.

This might be an omission, indeed.

> -Keyboard-only or keyboard dominant interface. This really ties in
> with the ASCII display in that roguelikes have their roots in console
> applications.

The ASCII (and the keyboard) points are not really important. At the time
those games were invented, they were the only choices available, but
mentioning it now feels a bit superfluous. Transferring a given roguelike
from keyboard/ASCII to mouse/tiles should affect the gameplay even less
than altering the setting.

Thanks for reading the pamphlet :)
David

Derek Ray

unread,
Sep 27, 2008, 9:10:52 PM9/27/08
to
On 2008-09-28, dpeg <pl...@zio.mathematik.hu-berlin.de> wrote:

> Darren Grey wrote:
>> -Character creation that allows numerous very different starting
>> characters (usually through a race/class system). This ties in with
>> the random generation to increase replayability, since it allows very
>> different games each playthrough.
> This might be an omission, indeed.

Note that the original canon, Rogue, does not have this.

--
Derek

Game info and change log: http://sporkhack.com
Beta Server: telnet://sporkhack.com
IRC: irc.freenode.net, #sporkhack

David Damerell

unread,
Sep 29, 2008, 11:03:52 AM9/29/08
to
Quoting dpeg <pl...@zio.mathematik.hu-berlin.de>:

>Darren Grey wrote:
>>-Computer RPG gameplay elements. In particular stats such as HP,
>>strength, etc, and levels/experience gained through killing monsters.
>We alluded to this in the last point: numbers. I think that's enough.

But not to character improvement, which _is_ a very common element, and
I'm surprised it's not on the list. Ali Harlow and I argued about this one
London RGRN, and I found it very hard to come up with a sensible idea for
a RL that didn't have character development.

>Again contrary (to your point), it might have been pointed out better that
>roguelikes are _not_ roleplaying games, I think :)

I know it's my pet gripe but it's clear, surely, that Darren Grey meant
"computer RPG" in the usual terminological muddle - "stats and levels and
sub-D&D mechanics", not "a roleplaying game"?
--
David Damerell <dame...@chiark.greenend.org.uk> Kill the tomato!
Today is First Brieday, September.

David Damerell

unread,
Sep 29, 2008, 11:13:25 AM9/29/08
to
Quoting Jeff Lait <torespon...@hotmail.com>:
>The game world is randomly generated in a way that increases
>replayability. Appearance and placement of items is random.
>Appearance of monsters is fixed, their placement is random.

I don't think a game with random monster properties would be less
roguelike for having them. Thematically horrible, maybe?

>Fixed content (plots or puzzles or vaults) removes randomness.

A nod to Gearhead?

>The world is represented by a uniform grid of tiles. Monsters (and
>the player) take up one tile, regardless of size.

NetHack's long worms don't really render it less roguelike...

>Even though there can be much more to the game, killing lots of
>monsters is a very important part of a roguelike. The game is player-
>vs-world: there are no monster/monster relations (like enmities, or
>diplomacy).

This I don't see. Oh, sure, the killing lots of monsters, at least until
someone comes up with something better - but lots of games have peaceful
and tame monsters, and quite a few have monster-monster interactions of
some kind - and I don't see that it would make a roguelike less roguelike.

>The traditional display for roguelikes is to represent the tiled world
>by ASCII characters.

I think a better criterion would be that it _could_ be so represented,
whether or not is _is_.

I think of randomness/turn-based/grid world/tactical learning/character
advancement as more fundamental.

dpeg

unread,
Sep 29, 2008, 2:47:35 PM9/29/08
to
David Damerell wrote:

> Quoting Jeff Lait <torespon...@hotmail.com>:
>>The game world is randomly generated in a way that increases
>>replayability. Appearance and placement of items is random.
>>Appearance of monsters is fixed, their placement is random.
>
> I don't think a game with random monster properties would be less
> roguelike for having them. Thematically horrible, maybe?

The idea is not to rule out any games that stray from the path. The
point is that all games in the canon use this approach to monsters.
(Yes, they also all have a kind of fantasy setting. This is not
crucial in my opinion. Their common mechanic of dealing with items
and monsters is.)



>>Fixed content (plots or puzzles or vaults) removes randomness.
>
> A nod to Gearhead?

No. Nods to ADOM (plot), Crawl (vaults) and Nethack (puzzles). I
believe that a game can gain from plots, puzzles, vaults. But
randomisation is a must, I think. (Randomisation can also come from
simply providing many miniplots adding up to one overarching story,
or many different vaults etc.)

>>The world is represented by a uniform grid of tiles. Monsters (and
>>the player) take up one tile, regardless of size.
>
> NetHack's long worms don't really render it less roguelike...

We didn't want to point each and every exception. Nethack follows the
system described to an extremely large extent. Also, r.g.r.n is full
enough of "why are (alive!) dragons and (alive!) giant newts of the
same size?" anyway.


>>Even though there can be much more to the game, killing lots of
>>monsters is a very important part of a roguelike. The game is player-
>>vs-world: there are no monster/monster relations (like enmities, or
>>diplomacy).
>
> This I don't see. Oh, sure, the killing lots of monsters, at least until
> someone comes up with something better - but lots of games have peaceful
> and tame monsters, and quite a few have monster-monster interactions of
> some kind - and I don't see that it would make a roguelike less roguelike.

You seem to have missed (or misunderstood) the preamble. It is definitely
possible to create roguelikes without focusing on killing. But all games
in the canon do. Most allies are just another tool for killing; and neutral
monsters are a very small minority in any case.

>>The traditional display for roguelikes is to represent the tiled world
>>by ASCII characters.
>
> I think a better criterion would be that it _could_ be so represented,
> whether or not is _is_.

This is what I suggested :)
But the others correctly pointed out that the central idea of your (or my)
phrase is grid-based gameplay, which is already covered above.

> I think of randomness/turn-based/grid world/tactical learning/character
> advancement as more fundamental.

Randomness, turn-based, grid-based are very high on the list. Tactical
learning might be a bit understated; character advancement is largely
missing (but a valid point). I think that permadeath is in some form pretty
crucial. Of course there is room for improvement. Something to do next
year :)

David

Ray Dillinger

unread,
Sep 29, 2008, 5:57:19 PM9/29/08
to
David Damerell wrote:


> I know it's my pet gripe but it's clear, surely, that Darren Grey meant
> "computer RPG" in the usual terminological muddle - "stats and levels and
> sub-D&D mechanics", not "a roleplaying game"?

David. Please don't go there again. We're all aware that
you don't think any single-player game can be an RPG. Fine.
Let's not waste a lot of attention and annoyance talking
about it.

Bear

David Damerell

unread,
Sep 30, 2008, 11:23:08 AM9/30/08
to
Quoting Ray Dillinger <be...@sonic.net>:
>David Damerell wrote:
>>I know it's my pet gripe but it's clear, surely, that Darren Grey meant
>>"computer RPG" in the usual terminological muddle - "stats and levels and
>>sub-D&D mechanics", not "a roleplaying game"?
>David. Please don't go there again.

FFS, try actually reading what I wrote. dpeg went there and I (unusually)
suggested in this case it wasn't necessary because it was obvious from
context.


--
David Damerell <dame...@chiark.greenend.org.uk> Kill the tomato!

Today is First Gouday, September.

David Damerell

unread,
Sep 30, 2008, 11:16:07 AM9/30/08
to
Quoting dpeg <pl...@zio.mathematik.hu-berlin.de>:
>David Damerell wrote:
>>Quoting Jeff Lait <torespon...@hotmail.com>:
>>>The game world is randomly generated in a way that increases
>>>replayability. Appearance and placement of items is random.
>>>Appearance of monsters is fixed, their placement is random.
>>I don't think a game with random monster properties would be less
>>roguelike for having them. Thematically horrible, maybe?
>The idea is not to rule out any games that stray from the path. The
>point is that all games in the canon use this approach to monsters.
>(Yes, they also all have a kind of fantasy setting. This is not
>crucial in my opinion. Their common mechanic of dealing with items
>and monsters is.)

I think it's incumbent on you to explain why a fantasy setting is not
crucial even though it is nearly ubiquitous (and I agree it is not
crucial) but fixed monster properties are crucial given that ubiquity is
not a sufficient criterion.

For me, randomness is indeed part of what makes roguelikes roguelikes; if
I sit down and imagine a game rather like a roguelike but where monster
properties are generated randomly but coherently, it still seems like that
game would be a roguelike.

>>>Fixed content (plots or puzzles or vaults) removes randomness.
>>A nod to Gearhead?
>No. Nods to ADOM (plot), Crawl (vaults) and Nethack (puzzles).

I mean that Gearhead has a plot but it is not fixed.

>>This I don't see. Oh, sure, the killing lots of monsters, at least until
>>someone comes up with something better - but lots of games have peaceful
>>and tame monsters, and quite a few have monster-monster interactions of
>>some kind - and I don't see that it would make a roguelike less roguelike.
>You seem to have missed (or misunderstood) the preamble. It is definitely
>possible to create roguelikes without focusing on killing. But all games
>in the canon do.

But, again, given the discarding of fantasy settings, merely because a
property is commonplace doesn't mean it's vital.


--
David Damerell <dame...@chiark.greenend.org.uk> Kill the tomato!

Today is First Gouday, September.

David Ploog

unread,
Sep 30, 2008, 11:56:37 AM9/30/08
to
On Tue, 30 Sep 2008, David Damerell wrote:
> Quoting dpeg <pl...@zio.mathematik.hu-berlin.de>:
>> David Damerell wrote:
>>> Quoting Jeff Lait <torespon...@hotmail.com>:

>>>> The game world is randomly generated in a way that increases
>>>> replayability. Appearance and placement of items is random.
>>>> Appearance of monsters is fixed, their placement is random.

>>> I don't think a game with random monster properties would be less
>>> roguelike for having them. Thematically horrible, maybe?

>> The idea is not to rule out any games that stray from the path. The
>> point is that all games in the canon use this approach to monsters.
>> (Yes, they also all have a kind of fantasy setting. This is not
>> crucial in my opinion. Their common mechanic of dealing with items
>> and monsters is.)
>
> I think it's incumbent on you to explain why a fantasy setting is not
> crucial even though it is nearly ubiquitous (and I agree it is not
> crucial) but fixed monster properties are crucial given that ubiquity is
> not a sufficient criterion.

Sure, it is incumbent (had to look that one up). I guess that a mentioning
of the fantasy setting in the minor category would have been fine with
everyone. To me, this is another instance where gameplay (how monsters are
canonically represented in roguelikes) is much more important than flavour
(that they are inspired by Tolkien or D&D).
Put another way, the "one monster per tile" may help in distinguishing the
genre (using this definition) from others, say strategy games (many of
which are turn-based, grid-based, randomly generated). What does the
mentioning of the fantasy trope bring?

> For me, randomness is indeed part of what makes roguelikes roguelikes; if
> I sit down and imagine a game rather like a roguelike but where monster
> properties are generated randomly but coherently, it still seems like that
> game would be a roguelike.

Sure, it would. As I see it, part of the definition game is to find out
which aspects of roguelikes could be changed in a fun way.

>>>> Fixed content (plots or puzzles or vaults) removes randomness.

> I mean that Gearhead has a plot but it is not fixed.

Ah, that sounds good, yes.

>>> This I don't see. Oh, sure, the killing lots of monsters, at least until
>>> someone comes up with something better - but lots of games have peaceful
>>> and tame monsters, and quite a few have monster-monster interactions of
>>> some kind - and I don't see that it would make a roguelike less roguelike.

>> You seem to have missed (or misunderstood) the preamble. It is definitely
>> possible to create roguelikes without focusing on killing. But all games
>> in the canon do.
>
> But, again, given the discarding of fantasy settings, merely because a
> property is commonplace doesn't mean it's vital.

Exactly. And just like designing means taking responsibility, we found
mentioning the hack'n slash important. Looking at it again from the
genre-classifying angle, this point rules out quite some types of games
(which are focused on economy, for example, or also puzzles).

David

Ray Dillinger

unread,
Sep 30, 2008, 1:09:36 PM9/30/08
to
dpeg wrote:

> Darren Grey wrote:

>> So can we start a flamewar now? ;)
>
> Of course! :)

> We tried to nail down properties
> related to gameplay, and/or dealing with differentiating from other
> genres. (And it might have been useful to state that.)

FWIW, I think trying to pin down an exact definition of a
roguelike game is a lot like trying to nail jelly to a tree.

I can't imagine any good reason for people at the gathering
in Berlin (or here for that matter) to spend time on it; it's
known to be unsolvable, why not just leave it alone?

Bear

Krice

unread,
Sep 30, 2008, 1:56:32 PM9/30/08
to
On 30 syys, 20:09, Ray Dillinger <b...@sonic.net> wrote:
> I can't imagine any good reason for people at the gathering
> in Berlin (or here for that matter) to spend time on it; it's
> known to be unsolvable, why not just leave it alone?

It's not hard at all. Roguelikes are basically computer
role-playing games with random content, but then defining
what a role-playing game is, that's the difficult question,
because there is quite big variation in games that have
"role-playing" features.

David Ploog

unread,
Sep 30, 2008, 2:01:57 PM9/30/08
to
On Tue, 30 Sep 2008, Ray Dillinger wrote:

> dpeg wrote:

>> We tried to nail down properties
>> related to gameplay, and/or dealing with differentiating from other
>> genres. (And it might have been useful to state that.)
>
> FWIW, I think trying to pin down an exact definition of a
> roguelike game is a lot like trying to nail jelly to a tree.
>
> I can't imagine any good reason for people at the gathering
> in Berlin (or here for that matter) to spend time on it; it's
> known to be unsolvable, why not just leave it alone?

You can easily see it that way. I think that becoming more conscious about
the crucial (and not so crucial) aspects of the genre is worthwhile. In
other forms of art (say music or architecture), people would likewise
debate about genres, without any definite outcome, and still not in vain.

David

dominik...@gmail.com

unread,
Sep 30, 2008, 5:42:16 PM9/30/08
to

I know it's a bit off-topic, but I don't quite feel like opening a new
thread just for this. Would you Berlin guys be so kind and upload the
videos to Youtube? I really find it a pain in the arse to look for
software capable of reproducing ogg video. Or to install it
exclusively to view the Berlin videos and uninstall it afterwards...

Mingos.

Jeff Lait

unread,
Sep 30, 2008, 9:05:16 PM9/30/08
to

Christopher Brandt is transcoding/editing the videos into more
friendly formats. I believe he plans as well on getting them into 10
minute chunks for YouTube. The initial ogm files are there to ensure
we have *something* available immediately.

If nothing gets done by 2009, I may have some free time to transcode
them myself :>
--
Jeff Lait
(POWDER: http://www.zincland.com/powder)

Jeff Lait

unread,
Sep 30, 2008, 9:07:25 PM9/30/08
to
On Sep 29, 11:03 am, David Damerell <damer...@chiark.greenend.org.uk>
wrote:

> Quoting dpeg <pl...@zio.mathematik.hu-berlin.de>:
>
> >Darren Grey wrote:
> >>-Computer RPG gameplay elements. In particular stats such as HP,
> >>strength, etc, and levels/experience gained through killing monsters.
> >We alluded to this in the last point: numbers. I think that's enough.
>
> But not to character improvement, which _is_ a very common element, and
> I'm surprised it's not on the list. Ali Harlow and I argued about this one
> London RGRN, and I found it very hard to come up with a sensible idea for
> a RL that didn't have character development.

I think I tended to subsume that in the Hack and Slash category - ie,
the idea that it is the standard hack-and-slash D&D campaign. On
closer look, you are right that it isn't there explicitly, so I think
this is a valid criticism of the list we finalized.

Jeff Lait

unread,
Sep 30, 2008, 9:24:27 PM9/30/08
to
On Sep 29, 11:13 am, David Damerell <damer...@chiark.greenend.org.uk>
wrote:

> Quoting Jeff Lait <torespondisfut...@hotmail.com>:
>
> >The game world is randomly generated in a way that increases
> >replayability. Appearance and placement of items is random.
> >Appearance of monsters is fixed, their placement is random.
>
> I don't think a game with random monster properties would be less
> roguelike for having them. Thematically horrible, maybe?

Poor editing in transferring our discussions to paper, I think :>
Those two lines were examples of *how* the random generation is
commonly done. Thus, stating that appearance of monsters is fixed
wasn't, in my mind, to suggest that unfixed monsters would be the
least bit unroguelike, but merely descriptive of the techniques in
rogue et al. The second half of that last sentence was the critical
part. We wanted to make it clear that *despite* Rogue having fixed
monster appearance, the location of the monsters is varied which leads
to this sort of replayability effect. Constrast with Zelda where each
room starts with the monsters in the same locations.

> >The world is represented by a uniform grid of tiles. Monsters (and
> >the player) take up one tile, regardless of size.
>
> NetHack's long worms don't really render it less roguelike...

Long worms are the exception that prove the rule.

> >Even though there can be much more to the game, killing lots of
> >monsters is a very important part of a roguelike. The game is player-
> >vs-world: there are no monster/monster relations (like enmities, or
> >diplomacy).
>
> This I don't see. Oh, sure, the killing lots of monsters, at least until
> someone comes up with something better - but lots of games have peaceful
> and tame monsters, and quite a few have monster-monster interactions of
> some kind - and I don't see that it would make a roguelike less roguelike.

There was blood spilled over this entry. I understand exactly what
you mean and argued quite vociferously to that effect. The trick to
understanding the intention of this is to step away from the view
point of the @ and look at it from the game designer. In this sense
the monsters are designed around combating the @. "Friends" and
"allies" are resources equipped by the @ against them - much like
scrolls and weapons are. The real litmus test is that monster/monster
combat tends to detract from a roguelike. My own experience with
POWDER agrees with this - the history of it is one of tightening up
monsters to have them interact less with each other and more against
the @. This ties to the game being a dungeon-crawl rather than a role
playing game.

Jeff Lait

unread,
Sep 30, 2008, 9:35:09 PM9/30/08
to
On Sep 30, 1:09 pm, Ray Dillinger <b...@sonic.net> wrote:
> dpeg wrote:
> > Darren Grey wrote:
> >> So can we start a flamewar now? ;)
>
> > Of course! :)
> > We tried to nail down properties
> > related to gameplay, and/or dealing with differentiating from other
> > genres. (And it might have been useful to state that.)
>
> FWIW, I think trying to pin down an exact definition of a
> roguelike game is a lot like trying to nail jelly to a tree.

Good thing we've given up on exact definitions :>

> I can't imagine any good reason for people at the gathering
> in Berlin (or here for that matter) to spend time on it; it's
> known to be unsolvable, why not just leave it alone?

At the Berlin meeting we had a series of presentations on roguelikes.
As such, I think it is pretty important that we have some communal
idea of what "roguelike" is. It seems more silly to have meetings
like that, or newsgroups called rec.games.roguelike, without any
understanding of what the term means.

I think roguelike studies has as its proper focus the analysis of
roguelikes to determine what makes this genre. How else can one
properly talk about how something is roguelike, or escapes the limits
of roguelikeness? Further, I think the game design of roguelikes
themselves can be informed by this sort of genre. It is very useful
for designers to know and understand why a grid based layout is
different from a continuous field.

Slash

unread,
Oct 1, 2008, 7:55:06 AM10/1/08
to
On Sep 26, 3:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
> ==Preamble==
>
> This definition of "Roguelike" was created at the International
> Roguelike Development Conference 2008 and is the product of a
> discussion between all who attended. The definition athttp://www.roguetemple.com/roguelike-definition/ was used as the

> starting point for the discussions.  Most factors are newly phrased,
> new factors have been added, some factors have been removed.

And some factors will be rephrased there as well, based on this
"Interpretation"

SNIP

> ==High value factors==
>
> ====Random environment generation====
>
> The game world is randomly generated in a way that increases
> replayability. Appearance and placement of items is random.
> Appearance of monsters is fixed, their placement is random.
> Fixed content (plots or puzzles or vaults) removes randomness.

Fixed overworld also removes randomness

> ====Permadeath====
>
> You are not expected to win the game with your first character.  You
> start over from the first level when you die.  (It is possible to save
> games but the savefile is deleted upon loading.)  The random
> environment makes this enjoyable rather than punishing.

This I already talked with Jeff about; I think the concept is
Permafailure, as it stands on the roguetemple definition. Permadeath
is just one kind of permafailure, roguelikes punish the partial and
not deadly failures of the player, like eating a long sword thinking
it would grant you sword breating instead of cutting your stomach.

SNIP

>
> ====Grid-based====
>
> The world is represented by a uniform grid of tiles.  Monsters (and
> the player) take up one tile, regardless of size.

What is a tile, and what is uniform? is a hex "grid" uniform or are we
restricting ourselves to cartesian spaces? Also, where are the items
here? must we say they also occupy a tile? should we add that they
must stack? (based on the choosen "canon")

> ====Non-modal====
>
> Movement, battle and other actions take place in the same mode.  Every
> action should be available at any point of the game. Violations to
> this are ADOM's overworld or Angand's and Crawl's shops.

Non-modal sounds like you can do things in parallel while playing a
roguelike (in contrast of being modal) :)

SNIP

>
> ====Exploration and discovery====
>
> The game requires careful exploration of the dungeon levels and
> discovery of the usage of unidentified items. This has to be done anew
> every time the player starts a new game.

Do they? what about power diving?

> ==Low value factors==
>
SNIP


>
> ====Dungeons====
>
> Roguelikes contain dungeons, such as levels composed of rooms and
> corridors.

Guess so, because of the "canon" games, but I would remove the theme
part from them and take only the abstract concepts

>
> ====Numbers====
>
> The numbers used to describe the character (hit points, attributes
> etc.) are deliberately shown.

Again, this is shared by all of the canonical games so I guess it is
ok


--
Slashie
http://slashie.net
http://roguetemple.com
http://santiagoz.com/blog

David Damerell

unread,
Oct 1, 2008, 9:50:43 AM10/1/08
to
Quoting Slash <java....@gmail.com>:

>On Sep 26, 3:13=A0pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
>>The game requires careful exploration of the dungeon levels and
>>discovery of the usage of unidentified items. This has to be done anew
>>every time the player starts a new game.
>Do they? what about power diving?

Depth-first exploration is still exploration, and it certainly wants to be
done carefully.


--
David Damerell <dame...@chiark.greenend.org.uk> Kill the tomato!

Today is First Chedday, September - a public holiday.

Jeff Lait

unread,
Oct 1, 2008, 10:23:22 AM10/1/08
to
On Oct 1, 7:55 am, Slash <java.ko...@gmail.com> wrote:
> On Sep 26, 3:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
>
> > ==Preamble==
>
> > This definition of "Roguelike" was created at the International
> > Roguelike Development Conference 2008 and is the product of a
> > discussion between all who attended. The definition athttp://www.roguetemple.com/roguelike-definition/ was used as the
> > starting point for the discussions. Most factors are newly phrased,
> > new factors have been added, some factors have been removed.
>
> And some factors will be rephrased there as well, based on this
> "Interpretation"

If you could archive the original somewhere rather than just editing
in place it would be best - otherwise the Berlin Interpretation will
be referencing the wrong version :>

> > ==High value factors==
>
> > ====Random environment generation====
>
> > The game world is randomly generated in a way that increases
> > replayability. Appearance and placement of items is random.
> > Appearance of monsters is fixed, their placement is random.
> > Fixed content (plots or puzzles or vaults) removes randomness.
>
> Fixed overworld also removes randomness

Yes. More importantly, fixed overworld often violates the Non-
Modality and the Uniform Grid terms.

> > ====Permadeath====
>
> > You are not expected to win the game with your first character. You
> > start over from the first level when you die. (It is possible to save
> > games but the savefile is deleted upon loading.) The random
> > environment makes this enjoyable rather than punishing.
>
> This I already talked with Jeff about; I think the concept is
> Permafailure, as it stands on the roguetemple definition. Permadeath
> is just one kind of permafailure, roguelikes punish the partial and
> not deadly failures of the player, like eating a long sword thinking
> it would grant you sword breating instead of cutting your stomach.

I don't like Permafailure as it is unclear. An essential point is
that you *restart*. You can't just compound mistake on mistake,
eventually you reach a point of no return where DYWYPI? is the only
option. This is seen in Nethack when people try to use Explore mode
but find themselves unable to continue despite immortality.

If we were to broaden this term, I'd characterize it as "No save and
continue", or "No take backs". Moves once made are not negotiable.
I'd then clarify this means the game can be made unwinnable by bad
moves.

> > ====Grid-based====
>
> > The world is represented by a uniform grid of tiles. Monsters (and
> > the player) take up one tile, regardless of size.
>
> What is a tile, and what is uniform? is a hex "grid" uniform or are we
> restricting ourselves to cartesian spaces? Also, where are the items
> here? must we say they also occupy a tile? should we add that they
> must stack? (based on the choosen "canon")

Uniform is tiles of roughly equal size and equal distance from each
other. A hex grid is definitely uniform. Items are left unclassified
- the stacking or not isn't seen as an essential roguelike
characteristic.

The reason that the grid is essential is that it opens up a sort of
tactical play which isn't possible with other spatial
representations. The discrete nature lets you easily plan your moves
- you know exactly how far you are from that dragon, you can know if
the door can be fully blocked by your pet, etc.

> > ====Non-modal====
>
> > Movement, battle and other actions take place in the same mode. Every
> > action should be available at any point of the game. Violations to
> > this are ADOM's overworld or Angand's and Crawl's shops.
>
> Non-modal sounds like you can do things in parallel while playing a
> roguelike (in contrast of being modal) :)

Don't let Windows reprogram your definitions :>

Wikipedia to the rescue:
http://en.wikipedia.org/wiki/Mode_(computer_interface)

Quoting thereof:
"In user interface design, a mode is a distinct setting within a
computer program or any physical machine interface, in which the same
user input will produce perceived different results than it would in
other settings"

I guess the right negation would be "Modeless", but for some reason I
jumped on Non-Modal.

Modal dialog boxes are called such not because they block other input,
but because they switch you from your previous input mode to a new one
where only the dialog is active.

> > ====Dungeons====
>
> > Roguelikes contain dungeons, such as levels composed of rooms and
> > corridors.
>
> Guess so, because of the "canon" games, but I would remove the theme
> part from them and take only the abstract concepts

I'm not so confident that theme can be fully distanced from gameplay.
Part of being a roguelike is the theme.

> > ====Numbers====
>
> > The numbers used to describe the character (hit points, attributes
> > etc.) are deliberately shown.
>
> Again, this is shared by all of the canonical games so I guess it is
> ok

I think this point is trying to accent the role of roguelikes as
single-player-character fantasy tactical combat simulator dungeon
crawls, rather than role playing games.

Hidden stats may make for a fun game, but prevent you from playing the
game in a roguelike fashion.

Darren Grey

unread,
Oct 1, 2008, 1:38:26 PM10/1/08
to

To be frank the dungeon style of rooms and corridors is so
quintessentially roguelike that I feel it deserves a mention, even
though it's completely unnecessary in a roguelike. Anyone familiar
with the genre looking at an @ symbol on a map of rectangular rooms
connected together by single-width corridors will instantly recognise
it as a roguelike. No other image can be more roguelike in style.

> > ====Numbers====
> >
> > The numbers used to describe the character (hit points, attributes
> > etc.) are deliberately shown.
>
> Again, this is shared by all of the canonical games so I guess it is
> ok

Actually, I'm not so sure it's that important a statement. How many
computer games *don't* show numbers describing the characters? More
important is the mechanics behind the numbers, which are predominantly
"computer RPG" style (emphasis on the quotation marks!) and generally
D&D-influenced.

I think this shouldn't be understated because the ability to customise
and develop your character is central to roguelike gameplay. In a
world with random dungeons, items and monster placements you have one
element of control through choosing skills, stats, specialities, etc
and improving your character through gaining experience. It also aids
the replayability of roguelikes in that you can improve and empower
your character in different ways in different games. It also gives
numerous approaches to overcoming certain areas in the game (even if
it is just a case of magic is stronger in certain places, etc). As a
develop it's important to remember that your character progression
system is one of the biggest gameplay elements, often the most
important to the player.

--
Darren Grey

David Ploog

unread,
Oct 2, 2008, 6:29:58 AM10/2/08
to
On Wed, 1 Oct 2008, Slash wrote:

> On Sep 26, 3:13�pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:

>> ====Random environment generation====
>>
>> The game world is randomly generated in a way that increases
>> replayability. Appearance and placement of items is random.
>> Appearance of monsters is fixed, their placement is random.
>> Fixed content (plots or puzzles or vaults) removes randomness.
>
> Fixed overworld also removes randomness

Yes. Not even sure if we need to list them all, this should be obvious :)

>> ====Permadeath====
>>
>> You are not expected to win the game with your first character. �You
>> start over from the first level when you die. �(It is possible to save
>> games but the savefile is deleted upon loading.) �The random
>> environment makes this enjoyable rather than punishing.
>
> This I already talked with Jeff about; I think the concept is
> Permafailure, as it stands on the roguetemple definition. Permadeath
> is just one kind of permafailure, roguelikes punish the partial and
> not deadly failures of the player, like eating a long sword thinking
> it would grant you sword breating instead of cutting your stomach.

I understood this. I was the one removing the "permafailure", and I did so
for two reasons: the more elaborate explanation you give here was not
present when we discussed it. And while I know that some roguelikes don't
have proper permadeath (but rather some serious disadvantage if you want
to keep playing the character), all of the canon do.


>> ====Grid-based====
>>
>> The world is represented by a uniform grid of tiles. �Monsters (and
>> the player) take up one tile, regardless of size.
>
> What is a tile, and what is uniform?

Twodimensional periodic tiling of the plane. You could mention that the
rectangular one is used most often (and by all games in the canon) but
this is a point where I think abstracting is better.

> Also, where are the items here? must we say they also occupy a tile?
> should we add that they must stack? (based on the choosen "canon")

1. It is often questioned why there can be only monster per tile.
We address this by adding a sentence on that effect.
2. Items are less important. Also, they do usually stack, but it matters
much less to gameplay anway.

I don't think we have to point out all common aspects of the games in the
canon. In my opinion, the goal is to single out the most relevant ones.
(As I said before, the common fantasy flavour is not really relevant, in
my opinion. Likewise, the common approach to items is much less relevant
than the common approach to monsters. Opinions will and must differ :)


>> ====Non-modal====
>>
>> Movement, battle and other actions take place in the same mode. �Every
>> action should be available at any point of the game. Violations to
>> this are ADOM's overworld or Angand's and Crawl's shops.
>
> Non-modal sounds like you can do things in parallel while playing a
> roguelike (in contrast of being modal) :)

We had "Uni-modal" before. I am not sure how to catch the idea in the best
way.

>>
>> ====Exploration and discovery====
>>
>> The game requires careful exploration of the dungeon levels and
>> discovery of the usage of unidentified items. This has to be done anew
>> every time the player starts a new game.
>
> Do they? what about power diving?

You still don't know how the dungeon looks like, so have to explore for
the stairs at least. I think the phrasing is fine.


>> ====Dungeons====
>>
>> Roguelikes contain dungeons, such as levels composed of rooms and
>> corridors.
>
> Guess so, because of the "canon" games, but I would remove the theme
> part from them and take only the abstract concepts

Yes, I can relate. But as far as I see it, 'dungeon' does not so much
refer to the flavour, but to the concept of 'rooms and corridors' (which
hints at tactical gameplay).

David

PS: Your definition was a great starting point. Given our approach, and
some of our omissions already pointed out, I can see slow convergence to a
nice and consistent pseudo-definition. Thank you!

Slash

unread,
Oct 2, 2008, 9:19:32 AM10/2/08
to
On Oct 2, 5:29 am, David Ploog <pl...@mi.fu-berlin.de> wrote:
> On Wed, 1 Oct 2008, Slash wrote:
> > On Sep 26, 3:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
> >> ====Random environment generation====
>
> >> The game world is randomly generated in a way that increases
> >> replayability. Appearance and placement of items is random.
> >> Appearance of monsters is fixed, their placement is random.
> >> Fixed content (plots or puzzles or vaults) removes randomness.
>
> > Fixed overworld also removes randomness
>
> Yes. Not even sure if we need to list them all, this should be obvious :)

Well, I propose "Fixed content (overworld, plots or puzzles or vaults)
removes randomness.", since the overworld is such an important thing
for living beings :P

> >> ====Permadeath====
>
> >> You are not expected to win the game with your first character.  You
> >> start over from the first level when you die.  (It is possible to save
> >> games but the savefile is deleted upon loading.)  The random
> >> environment makes this enjoyable rather than punishing.
>
> > This I already talked with Jeff about; I think the concept is
> > Permafailure, as it stands on the roguetemple definition. Permadeath
> > is just one kind of permafailure, roguelikes punish the partial and
> > not deadly failures of the player, like eating a long sword thinking
> > it would grant you sword breating instead of cutting your stomach.
>
> I understood this. I was the one removing the "permafailure", and I did so
> for two reasons: the more elaborate explanation you give here was not
> present when we discussed it. And while I know that some roguelikes don't
> have proper permadeath (but rather some serious disadvantage if you want
> to keep playing the character), all of the canon do.

My point is: permafailure includes permadeath, permadeath is thus a
high valued roguelikeness factor, but so are the rest of non-
recoverable failures (death is non-recoverable, but so are the rest of
non-favorable results caused by in-game actions); wearing a cursed
item without knowing it was cursed is a failure, and is permanent
(until you uncurse it, but you can't just roll it back like you would
in other genres)

SNIP

>
> > Also, where are the items here? must we say they also occupy a tile?
> > should we add that they must stack? (based on the choosen "canon")
>

SNIP

> 2. Items are less important. Also, they do usually stack, but it matters
>     much less to gameplay anway.

Why less important? I know the question of multi-tiles items is less
raised than the monsters counterpart, but items are an integral part
of the game world;; item stacking on the other hand may be a side
effect, a low level factor probably (it is a common thing, but
unneeded IMO)

As a side comment, I have heard it has some gameplay implications
(Polypiling?)

<b>How dares a crawl dev say items are less important! ^_^</b>

SNIP


>
> >> ====Dungeons====
>
> >> Roguelikes contain dungeons, such as levels composed of rooms and
> >> corridors.
>
> > Guess so, because of the "canon" games, but I would remove the theme
> > part from them and take only the abstract concepts
>
> Yes, I can relate. But as far as I see it, 'dungeon' does not so much
> refer to the flavour, but to the concept of 'rooms and corridors' (which
> hints at tactical gameplay).

I think this point is a bit weak... arent almost all top-down games
comprised of rooms and corridors? also, what about rooms only levels
(like ruins of buildings) or irregular cave levels?

>
> David
>
> PS: Your definition was a great starting point. Given our approach, and
> some of our omissions already pointed out, I can see slow convergence to a
> nice and consistent pseudo-definition. Thank you!

I'm glad it was useful :)

--
Slashie

Joe Hewitt

unread,
Oct 2, 2008, 10:10:14 AM10/2/08
to
There are some interesting points in this interpretation, and loads of
stuff to argue about. My first thought upon reading it was that it'd
be very interesting to create a 7DRL which attempts to follow this
definition exactly, taking its points to their logical extreme. My
second thought was that it would also be interesting to create a 7DRL
which tries to subvert the definition as much as possible while still
remaining within the spirit of the canon.

> ====Random environment generation====


> Fixed content (plots or puzzles or vaults) removes randomness.

Is it necessary to mention this as part of the definition? I will note
that every game in the canon, with the possible exception of Rogue,
has some degree of fixed content.

> ====Non-modal====
> Movement, battle and other actions take place in the same mode. Every
> action should be available at any point of the game. Violations to
> this are ADOM's overworld or Angand's and Crawl's shops.

This one interests me. It seems to me that nearly every Roguelike game
violates this point in some way. Does the inventory list count as a
separate mode? A yes/no prompt? How could you design a game to be
truly uni-modal while including content that normally gets shunted off
into a separate mode, and what would it look like? I think I have
another idea for a 7DRL.

- Joseph Hewitt
--
Comics> http://www.ataraxiatheatre.com/
GearHead> http://www.gearheadrpg.com/
GHForum> http://www.gearheadrpg.com/forum/

Radomir 'The Sheep' Dopieralski

unread,
Oct 2, 2008, 11:57:35 AM10/2/08
to
At Thu, 2 Oct 2008 07:10:14 -0700 (PDT),
Joe Hewitt wrote:

> There are some interesting points in this interpretation, and loads of
> stuff to argue about. My first thought upon reading it was that it'd
> be very interesting to create a 7DRL which attempts to follow this
> definition exactly, taking its points to their logical extreme. My
> second thought was that it would also be interesting to create a 7DRL
> which tries to subvert the definition as much as possible while still
> remaining within the spirit of the canon.

Yeah, that as my plan even before we started the discussion :)
But now I have lots of other ideas, so I guess I won't be a
competition for you.

>> ====Non-modal====
>> Movement, battle and other actions take place in the same mode. Every
>> action should be available at any point of the game. Violations to
>> this are ADOM's overworld or Angand's and Crawl's shops.
>
> This one interests me. It seems to me that nearly every Roguelike game
> violates this point in some way. Does the inventory list count as a
> separate mode? A yes/no prompt? How could you design a game to be
> truly uni-modal while including content that normally gets shunted off
> into a separate mode, and what would it look like? I think I have
> another idea for a 7DRL.

Actually z-day already does that. I've written it shortly after reading
Jeff Raskin's "Humane Interface". I'm not too pleased with the result,
but it might be just bad execution of a good idea.

--
Radomir Dopieralski

Antoine

unread,
Oct 2, 2008, 2:25:41 PM10/2/08
to
On Oct 3, 3:10 am, Joe Hewitt <pyrrh...@gmail.com> wrote:
> There are some interesting points in this interpretation, and loads of
> stuff to argue about. My first thought upon reading it was that it'd
> be very interesting to create a 7DRL which attempts to follow this
> definition exactly, taking its points to their logical extreme. My
> second thought was that it would also be interesting to create a 7DRL
> which tries to subvert the definition as much as possible while still
> remaining within the spirit of the canon.

Well, how could you do that?

A.

dpeg

unread,
Oct 2, 2008, 3:41:02 PM10/2/08
to
Slash wrote:

> On Oct 2, 5:29 am, David Ploog <pl...@mi.fu-berlin.de> wrote:
>> On Wed, 1 Oct 2008, Slash wrote:

>> >> ====Permadeath====
>>

[Snip definition of permadeath, and the proposal to use the term
permafailure instead.]

>> I understood this. I was the one removing the "permafailure", and I did
>> so for two reasons: the more elaborate explanation you give here was not
>> present when we discussed it. And while I know that some roguelikes don't
>> have proper permadeath (but rather some serious disadvantage if you want
>> to keep playing the character), all of the canon do.
>
> My point is: permafailure includes permadeath, permadeath is thus a
> high valued roguelikeness factor, but so are the rest of non-
> recoverable failures (death is non-recoverable, but so are the rest of
> non-favorable results caused by in-game actions); wearing a cursed
> item without knowing it was cursed is a failure, and is permanent
> (until you uncurse it, but you can't just roll it back like you would
> in other genres)

I find the term permafailure to be more technical and also more ambiguous.
Not being able to rewind is explicitly mentioned (you can save, but only
load once).


>> > Also, where are the items here? must we say they also occupy a tile?
>> > should we add that they must stack? (based on the choosen "canon")
>

>> 2. Items are less important. Also, they do usually stack, but it matters
>> much less to gameplay anway.
>
> Why less important? I know the question of multi-tiles items is less
> raised than the monsters counterpart, but items are an integral part
> of the game world; item stacking on the other hand may be a side
> effect, a low level factor probably (it is a common thing, but
> unneeded IMO)

It is a matter of taste, I guess. But to me, monsters are much more
important than items. I can easily imagine a roguelike without items; I
can't imagine one without monsters. Only one monster per tile seems worthy
of mention for several reasons (it is a deliberate departure from realism;
it distinguished the genre from others; it hints at the tactical aspect).
The items cannot compete in meaning. (I did mention the random appearance of
items to provide another example for randomisation, employed across the
canon.)

> <b>How dares a crawl dev say items are less important! ^_^</b>

Items rarely kill the player. Monsters do.


>> >> ====Dungeons====
>>
>> >> Roguelikes contain dungeons, such as levels composed of rooms and
>> >> corridors.
>>
>> > Guess so, because of the "canon" games, but I would remove the theme
>> > part from them and take only the abstract concepts
>>
>> Yes, I can relate. But as far as I see it, 'dungeon' does not so much
>> refer to the flavour, but to the concept of 'rooms and corridors' (which
>> hints at tactical gameplay).
>
> I think this point is a bit weak... arent almost all top-down games
> comprised of rooms and corridors? also, what about rooms only levels
> (like ruins of buildings) or irregular cave levels?

Strategy games are not. You could probably interpret this as a nod towards
computer RPGs.

David

dpeg

unread,
Oct 2, 2008, 3:43:00 PM10/2/08
to
Darren Grey wrote:

> On Oct 1, 12:55 pm, Slash <java.ko...@gmail.com> wrote:
>> On Sep 26, 3:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:

>> > ====Numbers====
>> >
>> > The numbers used to describe the character (hit points, attributes
>> > etc.) are deliberately shown.
>>
>> Again, this is shared by all of the canonical games so I guess it is
>> ok
>
> Actually, I'm not so sure it's that important a statement. How many
> computer games *don't* show numbers describing the characters? More
> important is the mechanics behind the numbers, which are predominantly
> "computer RPG" style (emphasis on the quotation marks!) and generally
> D&D-influenced.

Agreed. This point was added in a hurry. You are completely right on
character development (be that by attributes or skills or whatever).

Something to add for the next round :)

David

Joe Hewitt

unread,
Oct 2, 2008, 9:25:44 PM10/2/08
to

That's the challenge, isn't it? I guess I just have a bit of an
iconoclastic streak. Some of the items, such as "Appearance of
monsters is fixed", seem to be descriptions of what has come before
rather than prescriptions of what roguelikes should be, and so would
be easy to subvert. Others, such as permadeath, would obviously be
much harder (if not impossible) to subvert while remaining within the
spirit. The idea is to explore the definition by dealing with the
extreme cases.

This is an idea that just occurred off the top of my head- OlympusRL.
Have the PC be a god who is incapable of dying, but structure the game
in such a way that it's still quite easy to lose and there are still
many ways for the player to mess up. How would the mechanics of such a
system work? No idea. It's something to think about, though.

Darren Grey

unread,
Oct 3, 2008, 2:31:56 AM10/3/08
to
On Oct 3, 2:25 am, Joe Hewitt <pyrrh...@gmail.com> wrote:

> This is an idea that just occurred off the top of my head- OlympusRL.
> Have the PC be a god who is incapable of dying, but structure the game
> in such a way that it's still quite easy to lose and there are still
> many ways for the player to mess up. How would the mechanics of such a
> system work? No idea. It's something to think about, though.

I started doing something similar actually, whereby the player is
immortal, but certain obstacles could send it back to the beginning.
The gameplay itself is in a way puzzle-based - you have to use your
abilities (which could be developed in different directions) to
overcome the various obstacles. Of course in an abstract way regular
combat in roguelikes could be viewed as the same thing...
Replayability would be encouraged through an XBox style "Achievements"
system, which tries to get the player to win with the least mistakes,
the quickest times etc. In the end though I ditched the immortality
idea in all but plot because I figure the permadeath/failure is too
tied in with the random levels element to promote replayability.

--
Darren Grey

Ray Dillinger

unread,
Oct 3, 2008, 12:53:29 PM10/3/08
to
Darren Grey wrote:


> I started doing something similar actually, whereby the player is
> immortal, but certain obstacles could send it back to the beginning.
> The gameplay itself is in a way puzzle-based - you have to use your
> abilities (which could be developed in different directions) to
> overcome the various obstacles. Of course in an abstract way regular
> combat in roguelikes could be viewed as the same thing...
> Replayability would be encouraged through an XBox style "Achievements"
> system, which tries to get the player to win with the least mistakes,
> the quickest times etc. In the end though I ditched the immortality
> idea in all but plot because I figure the permadeath/failure is too
> tied in with the random levels element to promote replayability.

SmallGodsRL -- inspired by Terry Pratchett.

Your character is a god. A small god. If you put all your
effort and power into it, you just might be able to possess
an ant. This represents the pinnacle of your millennia of
existence, because recently you've had some luck; up to now
you've been strictly a god of amoebae. And if you think ants
are boring, you've never spent a thousand years trying to
attract the attention of an amoeba.

The game: Get noticed. Get honored. Get propitiated. Get
worshipped. Possess carefully selected creatures and do
carefully chosen things, solve problems or fulfill psychological
needs of those whose attention you want, and eventually attract
the attention of believers, which gives you power to do more.
Become a *real* god, a God of an intelligent race. If you
fail -- well, it's not death, but it's back to the amoebae
with all the quadrillions of other small gods....

Bear

Pointless

unread,
Oct 3, 2008, 9:48:38 PM10/3/08
to
I've tried to rephrase the original definitions to include some of the
suggestions in this thread. Keep in mind that this is merely my
opinion, and I am trying to integrate what I think are the most valid
points in the most forgiving way. Anything below may be used or edited
in any way.

On Sep 26, 4:13 pm, Jeff Lait <torespondisfut...@hotmail.com> wrote:
> ==Preamble==
>
> This definition of "Roguelike" was created at the International
> Roguelike Development Conference 2008 and is the product of a
> discussion between all who attended. The definition athttp://www.roguetemple.com/roguelike-definition/ was used as the
> starting point for the discussions.  Most factors are newly phrased,
> new factors have been added, some factors have been removed.
>

> ==General Principles==
>
> "Roguelike" refers to a genre, not merely "like-Rogue".  The genre is
> represented by its canon.  The canon for Roguelikes is ADOM, Angband,
> Crawl, Nethack, and Rogue.

add: For this reason, this definition can never define what a
Roguelike could be, but rather it is a cross-section of general
sentiment about what Roguelikes have in common.

> This list can be used to determine how roguelike a game is.  Missing
> some points does not mean the game is not a roguelike.  Likewise,
> possessing some points does not mean the game is a roguelike.
>
> The purpose of the definition is for the roguelike community to better
> understand what the community is studying.  It is not to place
> constraints on developers or games.
>
> ==High value factors==


>
> ====Random environment generation====
>
> The game world is randomly generated in a way that increases
> replayability. Appearance and placement of items is random.
> Appearance of monsters is fixed, their placement is random.

> Fixed content (plots or puzzles or vaults) removes randomness.

simplify: Randomization of game elements increases replayability.

> ====Permadeath====
>
> You are not expected to win the game with your first character.  You
> start over from the first level when you die.  (It is possible to save
> games but the savefile is deleted upon loading.)  The random
> environment makes this enjoyable rather than punishing.

change: Permafailure

Players must always suffer the consequences of their actions.
Restoring the game to a previous state is either strictly prohibited,
purposely difficult, or frowned upon by the Roguelike community. The


random environment makes this enjoyable rather than punishing.

The most common implementation of this is permadeath, where the game
must be started over upon player death.

> ====Turn-based====
>
> Each command corresponds to a single action/movement.  The game is not
> sensitive to time, you can take your time to choose your action.

Both of these are unclear. What if a command does not result in an
action? What if the game is turn-based but in real-time, like once a
second, a turn elapses, and your command is entered? What about combat
interruptions? I don't know how to save this one, but I feel that it
is essential in some way.

> ====Grid-based====
>
> The world is represented by a uniform grid of tiles.  Monsters (and
> the player) take up one tile, regardless of size.

simplify: The world is represented by a uniform grid of tiles, which
are occupied by items, monsters, characters, and other game world
objects.

> ====Non-modal====
>
> Movement, battle and other actions take place in the same mode.  Every
> action should be available at any point of the game. Violations to
> this are ADOM's overworld or Angand's and Crawl's shops.

change: Mode Reduction

There is a tendency to reduce the number of modal interfaces necessary
to interact with the game world. For example, combat occurs on the
same map as movement, without any scale change.

> ====Complexity====
>
> The game has enough complexity to allow several solutions to common
> goals. This is obtained by providing enough item/monster and item/item
> interactions and is strongly connected to having just one mode.

add: Many Roguelikes require specialized knowledge that comes from
repeated play attempts.

(Think about the canon and this will make sense)

> ====Resource management====
>
> You have to manage your limited resources (e.g. food, healing potions)
> and find uses for the resources you receive.

Good

> ====Hack'n'slash====


>
> Even though there can be much more to the game, killing lots of
> monsters is a very important part of a roguelike.  The game is player-
> vs-world: there are no monster/monster relations (like enmities, or
> diplomacy).

change last sentence: The game is player-vs-world, and characteristics
of monsters, the environment, items, and their relationships
completely revolve around creating an obstacle to player advancement.

> ====Exploration and discovery====
>
> The game requires careful exploration of the dungeon levels and
> discovery of the usage of unidentified items. This has to be done anew
> every time the player starts a new game.

Good. Possibly make the first "and" an "and/or" or simply an "or".

> ==Low value factors==

I like everything from this section except the last point:

> ====Numbers====
>
> The numbers used to describe the character (hit points, attributes
> etc.) are deliberately shown.

change (and possibly make a high-value factor): Character Advancement

The player character has persistent attributes that evolve throughout
the game. This is a central theme because the advancement of the
player character is usually essential to winning the game.

(This should absorb the RL-is-RPG complaints along with the RL-is-not-
RPG complaints in an intuitive way.)

dpeg

unread,
Oct 4, 2008, 9:47:30 PM10/4/08
to
Pointless wrote:

>> ====Random environment generation====
>>
>> The game world is randomly generated in a way that increases
>> replayability. Appearance and placement of items is random.
>> Appearance of monsters is fixed, their placement is random.
>> Fixed content (plots or puzzles or vaults) removes randomness.
>
> simplify: Randomization of game elements increases replayability.

I think that the specifications really add. Of course, your short conclusion
is correct, but the above hints at (the standard way of) how to get there.


>> ====Permadeath====
>>
>> You are not expected to win the game with your first character.  You
>> start over from the first level when you die.  (It is possible to save
>> games but the savefile is deleted upon loading.)  The random
>> environment makes this enjoyable rather than punishing.
>
> change: Permafailure
>
> Players must always suffer the consequences of their actions.
> Restoring the game to a previous state is either strictly prohibited,
> purposely difficult, or frowned upon by the Roguelike community. The
> random environment makes this enjoyable rather than punishing.
>
> The most common implementation of this is permadeath, where the game
> must be started over upon player death.

I agree that the original wording is clumsy, but I don't think the new
version makes it much clearer. First of all, "perma(nent)death" is so much
stronger than "perma(nent)failure" which is open to interpretation.


>> ====Grid-based====
>>
>> The world is represented by a uniform grid of tiles.  Monsters (and
>> the player) take up one tile, regardless of size.
>
> simplify: The world is represented by a uniform grid of tiles, which
> are occupied by items, monsters, characters, and other game world
> objects.

Mentioning that one tile can only inhibit one monster is useful. It gives a
distinction to (many) strategy games, for example.


>> ====Non-modal====
>>
>> Movement, battle and other actions take place in the same mode.  Every
>> action should be available at any point of the game. Violations to
>> this are ADOM's overworld or Angand's and Crawl's shops.
>
> change: Mode Reduction
>
> There is a tendency to reduce the number of modal interfaces necessary
> to interact with the game world. For example, combat occurs on the
> same map as movement, without any scale change.

I like this. I am completely confused by now on the title for this, though.
(One mode? Zero modes? Less modes? :)


>> ====Complexity====
>>
>> The game has enough complexity to allow several solutions to common
>> goals. This is obtained by providing enough item/monster and item/item
>> interactions and is strongly connected to having just one mode.
>
> add: Many Roguelikes require specialized knowledge that comes from
> repeated play attempts.
>
> (Think about the canon and this will make sense)

It does. If you have the canon point of view, the 'many' is not needed.


>> ====Hack'n'slash====
>>
>> Even though there can be much more to the game, killing lots of
>> monsters is a very important part of a roguelike.  The game is player-
>> vs-world: there are no monster/monster relations (like enmities, or
>> diplomacy).
>
> change last sentence: The game is player-vs-world, and characteristics
> of monsters, the environment, items, and their relationships
> completely revolve around creating an obstacle to player advancement.

This is okay, but hiding the fact that feuds/diplomacy is not used (in the
canon at least) makes is a bit more abstract.


>> ====Numbers====
>>
>> The numbers used to describe the character (hit points, attributes
>> etc.) are deliberately shown.
>
> change (and possibly make a high-value factor): Character Advancement
>
> The player character has persistent attributes that evolve throughout
> the game. This is a central theme because the advancement of the
> player character is usually essential to winning the game.
>
> (This should absorb the RL-is-RPG complaints along with the RL-is-not-
> RPG complaints in an intuitive way.)

This is a very good formulation. I'd just skip the 'usually'.

Theory is fun, eh? :)
David

Gerry Quinn

unread,
Nov 1, 2008, 5:34:03 PM11/1/08
to
In article <0763559a-b7a0-4da7-b82e-
c885e5...@d77g2000hsb.googlegroups.com>,
torespon...@hotmail.com says...

> ==Low value factors==
>
> ====Single player character====
>
> The player controls a single character. The game is player-centric,
> the world is viewed through that one character and that character's
> death is the end of the game.

I would put this as a high-value factor, multiple non-pet characters
tend to shade the game in the direction of "CRPG" IMO.

> ====ASCII display====
>
> The traditional display for roguelikes is to represent the tiled world
> by ASCII characters.

Replace "is" by "was". Graphic tiles are old and standard enough to
qualify as traditional now.

- Gerry Quinn
--
Lair of the Demon Ape (a coffee-break roguelike)
<http://indigo.ie/~gerryq/lair/lair.htm>

Gerry Quinn

unread,
Nov 1, 2008, 5:37:57 PM11/1/08
to
In article <gbmid7$ua5$02$1...@news.t-online.com>, pl...@zio.mathematik.hu-
berlin.de says...

> The ASCII (and the keyboard) points are not really important. At the time
> those games were invented, they were the only choices available, but
> mentioning it now feels a bit superfluous. Transferring a given roguelike
> from keyboard/ASCII to mouse/tiles should affect the gameplay even less
> than altering the setting.

Despite having made a mouse-based roguelike, I actually think that the
keyboard qualifies as a traditional and pretty standard roguelike
characteristic. Mouse-based control, unless it's merely a somewhat
inefficient way of implementing step-by-step movement, changes roguelike
combat tropes significantly.

Reply all
Reply to author
Forward
0 new messages