Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[ANN] New NetHack Variant: UnNetHack 3.5.1

164 views
Skip to first unread message

Patric Mueller

unread,
Jun 1, 2009, 12:46:29 PM6/1/09
to

Announcing the first development release of UnNetHack.


The main intent of this fork is to put more randomness, challenges and
fun into NetHack.


It features more levels (e.g. all of Pasi Kallinens new levels, most
of the Heck² levels by Kelly Bailey and the Blackmarket from
Slash'Em), several UI changes like automatically opened doors and the
pickup thrown patch and a lot of game play related changes.

All changes are documented on the trac project page.


On irc.freenode.net there's a channel #unnethack for discussion.


Trac project page:
http://apps.sourceforge.net/trac/unnethack-12312/

Source code and Win32 tty download:
http://sourceforge.net/project/showfiles.php?group_id=250747&package_id=306491&release_id=685462

Some screenshots:
http://apps.sourceforge.net/trac/unnethack-12312/wiki/Screenshots

David Ploog

unread,
Jun 1, 2009, 3:01:29 PM6/1/09
to
On Mon, 1 Jun 2009, Patric Mueller wrote:

> Announcing the first development release of UnNetHack.

> All changes are documented on the trac project page.

> http://apps.sourceforge.net/trac/unnethack-12312/

Just read through this, and it looks very good. I do not only mean the
various additions but also the nerfs and the overall design goals.

More power to you! I hope you cooperate with Derek or at least steal
what's good from Sporkhack :)
David

Serge in Darkness

unread,
Jun 2, 2009, 10:56:14 AM6/2/09
to
On 1 июн, 20:46, Patric Mueller <bh...@bigfoot.com> wrote:
> Trac project page:http://apps.sourceforge.net/trac/unnethack-12312/
>
> Source code and Win32 tty download:http://sourceforge.net/project/showfiles.php?group_id=250747&package_...
>
> Some screenshots:http://apps.sourceforge.net/trac/unnethack-12312/wiki/Screenshots

Do want PPC build. I cannot compile for the sake of my life.

Patric Mueller

unread,
Jun 2, 2009, 2:07:10 PM6/2/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
>
> Just read through this, and it looks very good. I do not only mean the
> various additions but also the nerfs and the overall design goals.

Thanks.

> More power to you! I hope you cooperate with Derek or

Not directly as UnNetHack up until know wasn't public and the design
goals are not completely the same.

But some SporkHack features have already found their way into
UnNetHack and I'm sure Derek won't hesitate to use the pieces
of UnNetHack that he likes.

A little more code exchange and more live variants like what happens
in the Angband world can't be bad for NetHack.

> at least steal what's good from Sporkhack :)

That's what public source code repositories are for :-)

Bye
Patric

--
NetHack-De: NetHack auf Deutsch - http://nethack-de.sourceforge.net/
NetHack for AROS: http://sourceforge.net/projects/nethack-aros/
UnNetHack: http://apps.sourceforge.net/trac/unnethack-12312/

Derek Ray

unread,
Jun 2, 2009, 4:42:24 PM6/2/09
to
On 2009-06-02, Patric Mueller <bh...@bigfoot.com> wrote:

> David Ploog <pl...@mi.fu-berlin.de> wrote:
>> More power to you! I hope you cooperate with Derek or
> But some SporkHack features have already found their way into
> UnNetHack and I'm sure Derek won't hesitate to use the pieces
> of UnNetHack that he likes.
>
> A little more code exchange and more live variants like what happens
> in the Angband world can't be bad for NetHack.

Yeah, this is pretty much my view of it too; if people like mine,
they'll play mine, if they like his, they'll play his, but if you get
into One Holy Branch syndrome you get problems like "5 years and
counting between bugfix releases."

--
Derek

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

David Ploog

unread,
Jun 4, 2009, 11:08:52 AM6/4/09
to
On Mon, 1 Jun 2009, Patric Mueller wrote:

> Announcing the first development release of UnNetHack.

I find it very bothersome that this announcement did not trigger any
feedback (apart from Derek of Spork and me, a renegade). As I see it,
there is a simple explanation:

a) The group is dead.
b) Only the devteam fan club is still here.

In either case, discussing conducts at length (conducts were introduced
because the game itself is too easy!) seems more interesting than
discussing a new variant.


Patric, you may already have heard it, but let me present an old idea and
observation of mine (I have been mentioning this three times in r.g.r.n
over the years, but I don't mind, since discussion here is pathetic
anyway):
The quest portal message and the quest entry screen are extremely
misleading. Both pretend there is some kind of urgency which does not
exist at all. Even worse, quest allies and monsters walk peacefully side
by side. Here are some ideas how to improve on that (I felt the
discrepancy between words and gameplay to be very disturbing when I was a
Nethack player):
a) The home level is war torn in some (ideally random) ways.
b) Use the grudge patch to make defenders and attackers battle each other.
This is not ideal because they will only fight when adjacent but it is
a start. Ideally, both attackers and defenders would go out of their
way to fight each other.
(On the quest leader: I suggest to open the stairs if he dies. There
should be some drawback to this procedure. One could also make him
invincible.)
c) Being quick to reach the quest leader should be worth something.
This can be done in two stages: be quick to find the portal. Be quick
on the portal level to reach the leader.
d) The more defenders survive, the better.

My sister and me wrote a patch many years ago. We've both moved to Crawl
since then but I can provide some more details, if you want.

David

Janis Papanagnou

unread,
Jun 4, 2009, 12:43:53 PM6/4/09
to
David Ploog wrote:
> On Mon, 1 Jun 2009, Patric Mueller wrote:
>
>> Announcing the first development release of UnNetHack.
>
> I find it very bothersome that this announcement did not trigger any
> feedback (apart from Derek of Spork and me, a renegade). As I see it,
> there is a simple explanation:
>
> a) The group is dead.
> b) Only the devteam fan club is still here.
>
> In either case, discussing conducts at length (conducts were introduced
> because the game itself is too easy!) seems more interesting than
> discussing a new variant.

I think discussing a variant requires that you've playtested that variant.
There certainly are people who play one game (or even more) per day, but
as far as I am concerned I finish just one game (or so) per month. Testing
every new variant with IMO marginal changes doesn't thrill me, to say the
least. There are a couple of good fixes and interesting changes in the two
latest variants that have been announced here, but just not enough changes
for me to interrupt my NAO documented sequence of games, or to spend more
of my time for yet another game - errm.. - another variant.

When I'm too tired to play myself but want to see some "NH(-variant)-TV"
I notice that on NAO there are mostly still vanilla games played, and the
Spork games, if any, either have stalled for a long period or there's just
one or two games trickling slowly on the first few dungeon levels. There
seems to be another server where the action happens? Or are the NAO players
just not interested in that variant? Sure we can speculate whether the game
(errm.., variant) isn't interesting or mature enough or whether your above
mentioned points a) or b) or maybe some c) is responsible for the lack of
interest there. Anyway. People talk/post/discuss about what they know and/
or about what they find interesting. Since folks can't know new variants
(without playtesting) those games require to trigger significant interest;
that seems not to be the case, yet. (And I am sure someone will point us
to some IRC/ICQ forum where you'll find vivid discussions for one of those
variants.)

Janis

David Ploog

unread,
Jun 4, 2009, 1:11:02 PM6/4/09
to
On Thu, 4 Jun 2009, Janis Papanagnou wrote:
> David Ploog wrote:
>> On Mon, 1 Jun 2009, Patric Mueller wrote:
>>
>>> Announcing the first development release of UnNetHack.
>>
>> I find it very bothersome that this announcement did not trigger any
>> feedback (apart from Derek of Spork and me, a renegade). As I see it, there
>> is a simple explanation:
>>
>> a) The group is dead.
>> b) Only the devteam fan club is still here.
>>
>> In either case, discussing conducts at length (conducts were introduced
>> because the game itself is too easy!) seems more interesting than
>> discussing a new variant.
>
> I think discussing a variant requires that you've playtested that variant.
> There certainly are people who play one game (or even more) per day, but
> as far as I am concerned I finish just one game (or so) per month. Testing
> every new variant with IMO marginal changes doesn't thrill me, to say the
> least. There are a couple of good fixes and interesting changes in the two
> latest variants that have been announced here, but just not enough changes
> for me to interrupt my NAO documented sequence of games, or to spend more
> of my time for yet another game - errm.. - another variant.

This essentially means you are forsaking development outside of the main
branch (with whatever development it has). I have seen discussions of
feature lists for new versions in various contexts, minutes after posting.
So I think your excuse is valid but it also reflects a certain
closed-mindedness.

> When I'm too tired to play myself but want to see some "NH(-variant)-TV"
> I notice that on NAO there are mostly still vanilla games played, and the
> Spork games, if any, either have stalled for a long period or there's just
> one or two games trickling slowly on the first few dungeon levels. There
> seems to be another server where the action happens? Or are the NAO players
> just not interested in that variant? Sure we can speculate whether the game
> (errm.., variant) isn't interesting or mature enough or whether your above
> mentioned points a) or b) or maybe some c) is responsible for the lack of
> interest there. Anyway. People talk/post/discuss about what they know and/
> or about what they find interesting. Since folks can't know new variants
> (without playtesting) those games require to trigger significant interest;
> that seems not to be the case, yet. (And I am sure someone will point us
> to some IRC/ICQ forum where you'll find vivid discussions for one of those
> variants.)

To me, it seems as if the lethargy of the devteam is infectious. Everybody
waiting for others to do something (and of course for the mythical new
OFFICIAL release). When somebody steps up and actually delivers something,
it is ignored -- on account of being too small (everyone starts small, and
if you've read UnNethack's changelog, you'd be surprised) or not popular
enough.

I am really happy that I am not part of this scene anymore and that
Crawl's playerbase is lively. This will come across as trolling but my
main point is that I feel so sorry for Patric.

David "Fremdschaemer"

Janis Papanagnou

unread,
Jun 4, 2009, 2:51:42 PM6/4/09
to
David Ploog wrote:
> On Thu, 4 Jun 2009, Janis Papanagnou wrote:
>> David Ploog wrote:
>>> On Mon, 1 Jun 2009, Patric Mueller wrote:
>>>
>>>> Announcing the first development release of UnNetHack.
>>>
>>> I find it very bothersome that this announcement did not trigger any
>>> feedback (apart from Derek of Spork and me, a renegade). As I see it,
>>> there is a simple explanation:
>>>
>>> a) The group is dead.
>>> b) Only the devteam fan club is still here.
>>>
>>> In either case, discussing conducts at length (conducts were
>>> introduced because the game itself is too easy!) seems more
>>> interesting than discussing a new variant.
>>
>> I think discussing a variant requires that you've playtested that
>> variant. There certainly are people who play one game (or even more) per
>> day, but as far as I am concerned I finish just one game (or so) per
>> month. Testing every new variant with IMO marginal changes doesn't thrill
>> me, to say the least. There are a couple of good fixes and interesting
>> changes in the two latest variants that have been announced here, but
>> just not enough changes for me to interrupt my NAO documented sequence of
>> games, or to spend more of my time for yet another game - errm.. -
>> another variant.
>
> This essentially means you are forsaking development outside of the main
> branch (with whatever development it has).

I don't see what makes you choose the word 'forsaking'. Is it because I
abstain from discussing changes of minor relevance? Or because I abstain
from re-discussing features that are incorporated in variants which have
already been extensively discussed before? Because I don't have the time
and interest to play a variant with IMO non-significant changes (compared
to the huge vanilla kernel)? <shrug>

> I have seen discussions of
> feature lists for new versions in various contexts, minutes after
> posting. So I think your excuse is valid but it also reflects a certain
> closed-mindedness.

Two points. First of all, my posting is no excuse; it explains why other
folks (like me) may not see "the problem" you see (whether existing or
not). Whether you consider it closed-mindedness is your choice; I think I
have just other priorities than you seem to have.

>> [...]


>
> To me, it seems as if the lethargy of the devteam is infectious.
> Everybody waiting for others to do something (and of course for the
> mythical new OFFICIAL release). When somebody steps up and actually
> delivers something, it is ignored -- on account of being too small
> (everyone starts small, and if you've read UnNethack's changelog, you'd
> be surprised) or not popular enough.

Again you seem to make wrong assumptions. The New Release won't come soon
(if at all). And I observed the variants and their features as close as
I've observed posted patches or (unimplemented) new ideas. So no, it is
not ignored, at least not by me.

> I am really happy that I am not part of this scene anymore and that
> Crawl's playerbase is lively. This will come across as trolling but my
> main point is that I feel so sorry for Patric.

Well, I don't think that he should feel bad about that. Patric seems to
me to be a decent person who is not putting his honorable work onto a
podest, and he is not talking and advertising his work with every posting
he makes, he is not hitchhiking threads to support acceptance, and he is
not ruling or implying that any discussion thread in RGRN is related to
UnNetHack. Frankly, from what I know about him (derived just from his
posting/cimmunication style, granted) I'd say he's a settled person, not
depending on public applause. (But who knows.)

Janis

Patric Mueller

unread,
Jun 5, 2009, 9:26:24 AM6/5/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
>
> To me, it seems as if the lethargy of the devteam is infectious.
> Everybody waiting for others to do something (and of course for the
> mythical new OFFICIAL release). When somebody steps up and actually

I think you've also seen the discussions in the last years where one
poster insisted that only a release blessed by the name of "NetHack"
would be considered official.

From those discussions I've concluded that the member of rgrn don't
agree and wouldn't be opposed to a new variant.

> delivers something, it is ignored -- on account of being too small
> (everyone starts small, and if you've read UnNethack's changelog,
> you'd be surprised) or not popular enough.

If the variant isn't popular enough even if it has the best changes
one can imagine then it just doesn't appeal to the players.

> I am really happy that I am not part of this scene anymore and that

I think you're still part of that community even if only as a
ex-player-with-such-a-strong-opinion-about-NetHacks-deficiencies-that-he-publicly-complains-about-a-IHO-not-enough-appreciated-variant-that-tries-to-fix-those-deficiencies
(lost a bit in translation, in German this would have been one word)
;-)

> Crawl's playerbase is lively. This will come across as trolling but my
> main point is that I feel so sorry for Patric.

You don't have to. :)

There are variants that have been longer around and would have
deserved more loving than UnNetHack. For example the biodiversity
patch, NetHack Brass and Lethe.

Especially Lethe is amazing. It solves the 'Gehennom is boring' in an
mind-blowing way by completely redesigning it with lots and lots of
cool levels.

> David "Fremdschaemer"

LOL

Patric Mueller

unread,
Jun 5, 2009, 9:51:48 AM6/5/09
to
Janis Papanagnou <janis_pa...@hotmail.com> wrote:
>
> I think discussing a variant requires that you've playtested that variant.
> There certainly are people who play one game (or even more) per day, but
> as far as I am concerned I finish just one game (or so) per month. Testing
> every new variant with IMO marginal changes doesn't thrill me, to say the
> least. There are a couple of good fixes and interesting changes in the two
> latest variants that have been announced here, but just not enough changes
> for me to interrupt my NAO documented sequence of games, or to spend more
> of my time for yet another game - errm.. - another variant.

I of course can't - and would want to - force anybody to try out
UnNetHack.

But just to put your argument regarding marginal changes into perspective:
SporkHack (which was announced almost 2 years ago) and UnNetHack
contain a lot more changes compared to 3.4.3 than NetHack 3.4.0 did in
comparison to 3.3.1.

Besides bug fixes and little tweaks the most obvious change has been
Kevin Bailey's patch for the Mines.


And considering minor changes I almost think the three best changes in
UnNetHack are UI changes:
- tty color scheme like Slash'Em
- auto opening of doors
- automatically pickup thrown objects (makes playing a dagger-thrower
fun)

David Ploog

unread,
Jun 5, 2009, 10:40:11 AM6/5/09
to
On Fri, 5 Jun 2009, Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:

>> I am really happy that I am not part of this scene anymore and that
>
> I think you're still part of that community even if only as a
> ex-player-with-such-a-strong-opinion-about-NetHacks-deficiencies-that-he-publicly-complains-about-a-IHO-not-enough-appreciated-variant-that-tries-to-fix-those-deficiencies
> (lost a bit in translation, in German this would have been one word)
> ;-)

Sure, I got carried away. What would the German word be? (You are correct
in your assessment, of course.) I do mean it however: if someone complains
in ##crawl about something that is a feature of Nethack, I always pointed
them to Spork. Now I'll point them to Spork and Un. I really believe in
cooperation and having more actively developed roguelikes to be very good
-- especially if they're as different as they happen to be. (Compare
POWDER, the active bands, Incursion, Nethack and Crawl, for example.)
However, Nethack is the odd one out in this list because it tries to avoid
all communication. This is why new variants are particularly useful there.


>> Crawl's playerbase is lively. This will come across as trolling but my
>> main point is that I feel so sorry for Patric.
>
> You don't have to. :)

Yes, that was me being bitter. I apologise.

> There are variants that have been longer around and would have
> deserved more loving than UnNetHack. For example the biodiversity
> patch, NetHack Brass and Lethe.
>
> Especially Lethe is amazing. It solves the 'Gehennom is boring' in an
> mind-blowing way by completely redesigning it with lots and lots of
> cool levels.

Do you plan to adopt (parts of) it?

David

Janis Papanagnou

unread,
Jun 5, 2009, 1:06:44 PM6/5/09
to
Patric Mueller wrote:
>
> I of course can't - and would want to - force anybody to try out
> UnNetHack.

Sure. I certainly didn't mean to say that you'd force anybody, rather
(I thought) I said the opposite; that you don't advocate your version
on every possible opportunity.

> But just to put your argument regarding marginal changes into perspective:
> SporkHack (which was announced almost 2 years ago) and UnNetHack
> contain a lot more changes compared to 3.4.3 than NetHack 3.4.0 did in
> comparison to 3.3.1.

True.

> [...]


> And considering minor changes I almost think the three best changes in
> UnNetHack are UI changes:

Yep, the NH UI required some improvement.

Janis

Doug Freyburger

unread,
Jun 5, 2009, 4:40:03 PM6/5/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
>
> To me, it seems as if the lethargy of the devteam is infectious. Everybody
> waiting for others to do something (and of course for the mythical new
> OFFICIAL release). When somebody steps up and actually delivers something,
> it is ignored -- on account of being too small (everyone starts small, and
> if you've read UnNethack's changelog, you'd be surprised) or not popular
> enough.

I continuing playing 3.1.1 for about 2 years after 3.4.3 was
released. Sporkhack has been out for 2 years so I'm due to
switch to it. Lethargy? Sure, or call it conservative or just
not in a hurry.

hcobb

unread,
Jun 5, 2009, 4:48:45 PM6/5/09
to
On Jun 1, 9:46 am, Patric Mueller <bh...@bigfoot.com> wrote:
> Announcing the first development release of UnNetHack.

What's the address to telnet in and test it out?

You know, just like sporkhack.com

No?

-HJC

Patric Mueller

unread,
Jun 5, 2009, 4:55:28 PM6/5/09
to

There's no public server yet.

But it is planned.

As soon as it goes live, it also will be announced here.

ran...@pactechdata.com

unread,
Jun 5, 2009, 9:55:33 PM6/5/09
to
On Jun 5, 6:51 am, Patric Mueller <bh...@bigfoot.com> wrote:
> And considering minor changes I almost think the three best changes in
> UnNetHack are UI changes:
>  - tty color scheme like Slash'Em
>  - auto opening of doors
>  - automatically pickup thrown objects (makes playing a dagger-thrower
>    fun)

The pickup-thrown patch as incorporated into slash'em has
a bug with thrown items that return (Mjollnir for Valks, and
boomerangs for anybody). The was-thrown bit sticks, so if you
later drop the item deliberately and then later still walk over
it, it will be auto-picked-up (sic) erroneously. I fixed that
for nethack by moving the place where the bit gets cleared from
the pickup code to the add-to-inventory code.

Jonker

unread,
Jun 5, 2009, 10:46:57 PM6/5/09
to
Patric Mueller wrote:
> Announcing the first development release of UnNetHack.
>
>
> The main intent of this fork is to put more randomness, challenges and
> fun into NetHack.

The whole reason I periodically check this newsgroup is in hopes of
happy surprises like this. Will follow your efforts. Keep up the good work!

a...@a.invalid

unread,
Jun 6, 2009, 12:44:53 AM6/6/09
to
Patric Mueller <bh...@bigfoot.com> wrote:

[...]


> Especially Lethe is amazing. It solves the 'Gehennom is boring' in an
> mind-blowing way by completely redesigning it with lots and lots of
> cool levels.

Indeed. In fact, I don't play NetHack since about 1999. I played Slash
and later the great Slash'em and then the Lethe patched Slash'em. There
is no way for me to 'go back' and plain vanilla NetHack at all.

But since some years now, I only play crawl or rather crawl-ref.

I love such games, I really do, having played or tested almost all there
are out there. ;)

Patric Mueller

unread,
Jun 6, 2009, 6:47:33 AM6/6/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
>
> The quest portal message and the quest entry screen are extremely
> misleading. Both pretend there is some kind of urgency which does not
> exist at all. Even worse, quest allies and monsters walk peacefully
> side by side. Here are some ideas how to improve on that (I felt the
> discrepancy between words and gameplay to be very disturbing when I
> was a Nethack player):

I've always thought that the player is late for the party.

The quest leader has posted the telepathic message and the magic
portal in the Dungeons of Doom as the quest nemesis was attacking the
home level.

So there is no more urgency when you arrive.

> a) The home level is war torn in some (ideally random) ways.

At least a few player corpses and items lying around that are fitting
to the players role would be nice.

The Archaeologist quest level has already a little bit of that.

> b) Use the grudge patch to make defenders and attackers battle each other.
> This is not ideal because they will only fight when adjacent but it is
> a start. Ideally, both attackers and defenders would go out of their
> way to fight each other.

In the usual quest level it looks like your people have locked
themselves in their buildings.

Isn't there the risk that the player just watches the fight without
having to do something?

> (On the quest leader: I suggest to open the stairs if he dies. There
> should be some drawback to this procedure. One could also make him
> invincible.)

SporkHack does open the stairs when the quest leader dies to alway
have the possibility to do the quest.

> c) Being quick to reach the quest leader should be worth something.
> This can be done in two stages: be quick to find the portal. Be quick
> on the portal level to reach the leader.

It should be worth something but what should that something be?

And it might even be harder to judge what's quick on the different
quest levels.

> d) The more defenders survive, the better.

How would that be accomplished that the player gets an advantage?

> My sister and me wrote a patch many years ago. We've both moved to Crawl
> since then but I can provide some more details, if you want.

Yes, please do, there are some things that are unclear to me how that
could be accomplished.

Patric Mueller

unread,
Jun 6, 2009, 8:31:45 AM6/6/09
to
ran...@pactechdata.com wrote:
>
> The pickup-thrown patch as incorporated into slash'em has
> a bug with thrown items that return (Mjollnir for Valks, and
> boomerangs for anybody). The was-thrown bit sticks, so if you
> later drop the item deliberately and then later still walk over
> it, it will be auto-picked-up (sic) erroneously.

I know of this bug and have a ticket for it on the trac page.

I decided that the patch with the bug is useful enough and postponed
fixing it.

> I fixed that
> for nethack by moving the place where the bit gets cleared from
> the pickup code to the add-to-inventory code.

Thanks for the tip. That'll save me some time.

Janis Papanagnou

unread,
Jun 6, 2009, 8:40:10 AM6/6/09
to
Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:
>
>> c) Being quick to reach the quest leader should be worth something.
>> This can be done in two stages: be quick to find the portal. Be quick
>> on the portal level to reach the leader.
>
> It should be worth something but what should that something be?
>
> And it might even be harder to judge what's quick on the different
> quest levels.

I think this is a bad idea. Velocity would depend on having or not having
equipment to detect the portal quickly. What about stumbling randomly into
the portal without being XL:14. Introducing timing effects in a game like
Nethack requires well thought out design; the above suggestion is not more
than a cursory YANI posting. Why, in the first place, should it "be worth
something"?

Janis

David Ploog

unread,
Jun 6, 2009, 9:11:03 AM6/6/09
to
On Sat, 6 Jun 2009, Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:
>>
>> The quest portal message and the quest entry screen are extremely
>> misleading. Both pretend there is some kind of urgency which does not
>> exist at all. Even worse, quest allies and monsters walk peacefully
>> side by side. Here are some ideas how to improve on that (I felt the
>> discrepancy between words and gameplay to be very disturbing when I
>> was a Nethack player):
>
> I've always thought that the player is late for the party.
>
> The quest leader has posted the telepathic message and the magic
> portal in the Dungeons of Doom as the quest nemesis was attacking the
> home level.
>
> So there is no more urgency when you arrive.

But on some (all?) quest levels the defenders still live and the attackers
still attack.

>> b) Use the grudge patch to make defenders and attackers battle each other.
>> This is not ideal because they will only fight when adjacent but it is
>> a start. Ideally, both attackers and defenders would go out of their
>> way to fight each other.
>
> In the usual quest level it looks like your people have locked
> themselves in their buildings.

It's years since I had a look at it, but I recall attackers and defenders
walk side by side in the open for the Valk quest.

> Isn't there the risk that the player just watches the fight without
> having to do something?

Sure. What I propose is not to punish those players, but to give some
reward to players who do better.

>> (On the quest leader: I suggest to open the stairs if he dies. There
>> should be some drawback to this procedure. One could also make him
>> invincible.)
>
> SporkHack does open the stairs when the quest leader dies to alway
> have the possibility to do the quest.

I like this.

>> c) Being quick to reach the quest leader should be worth something.
>> This can be done in two stages: be quick to find the portal. Be quick
>> on the portal level to reach the leader.
>
> It should be worth something but what should that something be?

For our patch years ago, my sister and I had two prizes (randomly chosen,
in Sokoban fashion). There is a score of your performance, taking into
account time spent between getting the telepathic message and entering the
portal; time spent between entering the portal and reaching the leader;
attackers killed; defenders lost. IIRC, score was a number between 0 and
10 or so.

a) The quest leader can improve your gear: he will try [score] times to
enchant your worn armour or wielded weapon(s). The lower chances for
already enchanted gear are applied, but he won't break anything.

b) The quest leader can create another portal from Home:1 to the dungeon.
The larger the score, the shallower that portal will be, thus creating a
larger shortcut in the dungeon graph. We got stuck with the portal (making
sure that pets follow etc.)

In hindsight, I still like a) but not b): the enchantment is interesting:
you can arrive in mundane gear and get a good boost or you can wield/wear
only a single item, trying to get past the safe enchantment threshold.

> And it might even be harder to judge what's quick on the different
> quest levels.

Yes, our formula for the score has a class-specific value: the ranger
quest is way slower than the wizard quest, of course.

>> d) The more defenders survive, the better.
>
> How would that be accomplished that the player gets an advantage?

The [score] went down for each defender killed.

>> My sister and me wrote a patch many years ago. We've both moved to Crawl
>> since then but I can provide some more details, if you want.
>
> Yes, please do, there are some things that are unclear to me how that
> could be accomplished.

I guess my main point is that the quest level and some features around it
are highly unrealistic (in an in-game sense). By actually making them
matter, it provides a nice tactical challenge: reaching the quest leader
asap is an interesting minigame, where wands, speed potions, jumping boots
and scrolls of stinking cloud can be put to good effect.

Cheers,
David

David Ploog

unread,
Jun 7, 2009, 2:29:29 AM6/7/09
to

You can get to the quest leader just fine regardless of XL. If he has to
tell you that you cannot tackle the quest proper right now, that is fine.
At least you proved that you care.

The proposal introduces a minigame; if you want to play it optimal, you
would make sure that you bring useful items when reaching the levels where
the portal can occur (wands of secret detection, magic mapping for the
portal hunt; various items for the quest leader hunt and still other items
for killing the quest attacks in quick ways).

Gameplay rationale: Having a racing type of challenge that early could set
another pre-Castle goalpost. You can freely choose to ignore the challenge
an wouldn't be worse off than in vanilla. But if you take it on, it can be
an interesting tactical operation.

Flavour rationale: There is urgency implied by the the first call. There
is urgency implied by the screenful when you go through the portal. There
is (or was) supposedly a war going on on Home:1. All of this is done via
messages and text, and the game does not live up to it all. So even if you
regard all of this as flavour text, it is bad flavour text.

(Note: I realise that the quests are not an urgent construction site for
Nethack. But I believe that the current system is unlogical, to the point
of being distorted, and could be improved in a rather simple manner.)

I also believe that you (Janis) are among the truest of the true. If I
wanted to picture a man who is the last worshipper of a religion, it
could be you :)

David

PS: I really love the "Introducing timing effects in a game like Nethack
requires well thought out design" bit. Nethack would be a different game
if someone had given deep thought to design and balance throughout. (I am
quite aware that some developers put a lot of effort into this in various
places. But then someone else came along and added a new shiny feature,
and the old balance was thrown out of the window.)

Patric Mueller

unread,
Jun 7, 2009, 6:50:27 AM6/7/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
> On Fri, 5 Jun 2009, Patric Mueller wrote:
>> David Ploog <pl...@mi.fu-berlin.de> wrote:
>
>>> I am really happy that I am not part of this scene anymore and that
>>
>> I think you're still part of that community even if only as a
>> ex-player-with-such-a-strong-opinion-about-NetHacks-deficiencies-that-he-publicly-complains-about-a-IHO-not-enough-appreciated-variant-that-tries-to-fix-those-deficiencies
>> (lost a bit in translation, in German this would have been one word)
>> ;-)
>
> Sure, I got carried away. What would the German word be? (You are

Ex-NetHack-Jammerspielerkritteler ;-)

> correct in your assessment, of course.) I do mean it however: if
> someone complains in ##crawl about something that is a feature of
> Nethack, I always pointed them to Spork. Now I'll point them to Spork

Do you have an example for that?

I'm still usually dying in the starting levels in Crawl, so this might
help me understand better the difference between Crawl and NetHack.

> and Un. I really believe in cooperation and having more actively
> developed roguelikes to be very good -- especially if they're as
> different as they happen to be. (Compare POWDER, the active bands,
> Incursion, Nethack and Crawl, for example.)

I'm still surprised how much tweaking the bands get, even though it
looks to me as an outsider that those tweaks don't change a lot. But
from the comments of the players it seems to make a difference.

> However, Nethack is the odd one out in this list because it tries to
> avoid all communication. This is why new variants are particularly
> useful there.

Avoiding *all* communication is probably a bit exaggerated. Of course
compared to the common OSS RL there is extremely little communication.

Why the DevTeam went into a almost complete seclusion while still
working (some call it 'claiming to work') on the next version nobody
knows.

>> There are variants that have been longer around and would have
>> deserved more loving than UnNetHack. For example the biodiversity
>> patch, NetHack Brass and Lethe.
>>
>> Especially Lethe is amazing. It solves the 'Gehennom is boring' in an
>> mind-blowing way by completely redesigning it with lots and lots of
>> cool levels.

I forgot that Lethe does have a rather peculiar feature. The filler
maze level of Gehennom can have instead of walls made of stone walls
made of lava/ice/moat/ironbars.

Those are quite ugly and that is unfortunate as Slash'Em development
got stuck halfway through integrating Lethe with monsters and those
filler levels integrated but the cool levels still out. :-(

> Do you plan to adopt (parts of) it?

I already have added the lethe levels from the Valley, Asmodeus,
Baalzebub and Orcus as alternatives to the ones from vanilla and
Cthulhu has taken the place of the high priest of Moloch as keeper of
the AoY.

I don't think that I'll add more major demons levels as the added
benefit of 6 instead of 4 special demons levels is small. But maybe
randomly choosing 4 out of 6?

Lethe has some beautiful level designs but also some levels are just
there without much use. For example the level where the river Lethe
starts.

I don't think the lower level need more branching.


NetHack Brass also has a nice design for the Gehennom levels:
http://www.maroon.dti.ne.jp/youkan/nethack/index_e.html

It divides Gehennom into four parts and each part leads to a special
level with the filler levels having a thematic look fitting to the
special level.

For example the filler level before Orcus' level are deserted towns
and those before Baalzebub's are cavern with forests of dead trees.

David Ploog

unread,
Jun 7, 2009, 11:55:50 AM6/7/09
to
On Sun, 7 Jun 2009, Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:
>> On Fri, 5 Jun 2009, Patric Mueller wrote:
>>> David Ploog <pl...@mi.fu-berlin.de> wrote:

>>> I think you're still part of that community even if only as a
>>> ex-player-with-such-a-strong-opinion-about-NetHacks-deficiencies-that-he-publicly-complains-about-a-IHO-not-enough-appreciated-variant-that-tries-to-fix-those-deficiencies
>>> (lost a bit in translation, in German this would have been one word)
>>> ;-)
>>
>> Sure, I got carried away. What would the German word be? (You are
>
> Ex-NetHack-Jammerspielerkritteler ;-)

I see. And agree :)

>> if someone complains in ##crawl about something that is a feature of
>> Nethack, I always pointed them to Spork. Now I'll point them to Spork
>
> Do you have an example for that?

##crawl is an IRC channel, so communication is not permanent. If you
really want me to, I can ask one of the guys who keeps ##crawl on his
harddisk. I also added an entry for Sporkhack in some roguelike listing,
including links.
It may come across differently, but my point is not at all to prove that
Nethack is a bad game (it is not!) or it should die out (it should not!).
My point is that Nethack is not finished and the devteam has probably
abandoned it (and even if not, they did not pay much attention to
balancing during the last releases that I've witnesses [*]). Therefore,
live variants should be encouraged, especially those that have an eye on
balance and global design. Thus, I would not point players in ##crawl to
Slashem.

[*] Yes, they've toned down riding after they introduced it. Which _is_ a
good case of balancing. But as far as I know, older problems have never
been addressed since 3.2.3.

> I'm still usually dying in the starting levels in Crawl, so this might
> help me understand better the difference between Crawl and NetHack.

Ah, you actually meant something different. Here are some typical comments
that prompted me to direct them to Spork:

Q: "There are so many unavoidable deaths. Wouldn't it be better if every
single game was winnable."
A: "Early game mortality is lower than you might expect. For a rogulike
designed around 100% winnability, go Spork".

Q: "Crawl is too random! Couldn't we have *some* granted loot at branch
ends?"
A: "Try Spork."

Q: "I want to be able to dip potions [or have otherwise more elaborate
item interactions."
A: "You know where to get Nethack but try Spork."

and some more.

Since you are a designer, you might just read the pas-de-faq (press ?n
when playing Crawl). Several of the points there address Nethack
indirectly. For last year's roguelike meeting, I wrote a small article on
design, you can read it here:
http://people.msoe.edu/~brayshaj/files/irdc/irdc.2008.David.Ploog.v2.pdf

>> Do you plan to adopt (parts of) Lethe?


>
> I already have added the lethe levels from the Valley, Asmodeus,
> Baalzebub and Orcus as alternatives to the ones from vanilla and
> Cthulhu has taken the place of the high priest of Moloch as keeper of
> the AoY.
>
> I don't think that I'll add more major demons levels as the added
> benefit of 6 instead of 4 special demons levels is small. But maybe
> randomly choosing 4 out of 6?

This may be my Crawl-brain replying, but I'd suggest to do that, yes.

> NetHack Brass also has a nice design for the Gehennom levels:
> http://www.maroon.dti.ne.jp/youkan/nethack/index_e.html
>
> It divides Gehennom into four parts and each part leads to a special
> level with the filler levels having a thematic look fitting to the
> special level.

I wasn't aware of NH Brass. This looks well thought out!

Cheers,
David

Janis Papanagnou

unread,
Jun 7, 2009, 1:13:17 PM6/7/09
to
David Ploog wrote:
>
> The proposal introduces a minigame; [...]

Sure. I like the idea of minigames in Nethack, and another and new one is
certainly worth discussing.

> if you want to play it optimal, you
> would make sure that you bring useful items when reaching the levels
> where the portal can occur (wands of secret detection, magic mapping for
> the portal hunt; various items for the quest leader hunt and still other
> items for killing the quest attacks in quick ways).

You imply (to solve the issue) existence of items that are not very common
to have on levels around 11-16, IME. You should also consider that you
need to be XL:14. Immediately doing the Quest, once on the portal level,
is untypical; usually I have to proceed much deeper (Castle/Valley) until
I am admitted. Also; some quests are far too challenging for many character
classes to be able to immediately enter them, you need to prepare for that.

> Gameplay rationale: Having a racing type of challenge that early could
> set another pre-Castle goalpost. You can freely choose to ignore the
> challenge an wouldn't be worse off than in vanilla. But if you take it
> on, it can be an interesting tactical operation.
>
> Flavour rationale: There is urgency implied by the the first call. There
> is urgency implied by the screenful when you go through the portal.
> There is (or was) supposedly a war going on on Home:1. All of this is
> done via messages and text, and the game does not live up to it all. So
> even if you regard all of this as flavour text, it is bad flavour text.

The text might not be optimal.

> (Note: I realise that the quests are not an urgent construction site for
> Nethack. But I believe that the current system is unlogical, to the
> point of being distorted, and could be improved in a rather simple manner.)

Hmm.. - Have you, maybe, any idea how to improve the Quest logic without
introducing timing effects? (Which would be the best way, IMO.) Or a way
to improve your idea so that the random differences, which are typically
huge between different instances of the game and between different classes
played, wouldn't make the "rewards" (whatever that would be) arbitrary.
I think answering those questions will make your "YANI" type of idea a
more serious feature.

> I also believe that you (Janis) are among the truest of the true. If I
> wanted to picture a man who is the last worshipper of a religion, it
> could be you :)

Interesting picture concerning someone who (not yet) believes in stories
from any religion, including the earthly "religions" (opinions) we are
told to believe every other day. ;-) But it is certainly correct that I
don't follow every new proposal or worship the ideology that *any* change
to the game is better than only sophisticated thought-out changes.

And you should be really careful with such statements; they don't make a
proposed idea better. Over the decades there's been a lot suggestions to
improve Nethack (I was no exception WRT making proposals), some were very
good, a lot more were just bad, and even amongst the good proposals were
many that disbalance the game, probably fix a minor issue but introduce a
larger problem, or make just no sense.

Are there any ideas, meanwhile, what the price could be? If the feature
will not be elaborated beyond the currently discussed state then I'd
favour _Score Points_ as reward, lots of Score Points! You know why :-p

Janis

Janis Papanagnou

unread,
Jun 7, 2009, 1:20:04 PM6/7/09
to
Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:
>>
>> b) Use the grudge patch to make defenders and attackers battle each other.
>> This is not ideal because they will only fight when adjacent but it is
>> a start. Ideally, both attackers and defenders would go out of their
>> way to fight each other.
>
> In the usual quest level it looks like your people have locked
> themselves in their buildings.
>
> Isn't there the risk that the player just watches the fight without
> having to do something?

The monster generation rate could be tuned (might be tricky!) to create
more foes if the player character doesn't engage in the defense activities
to decimate the foes. If the player avoids the fight more and more friendly
monsters would be killed and the foes would populate the quest levels. On
the way back the player would have to pass more monsters which might make
it more difficult, comparable to the crowd of inhabitants on Astral (but
surely not as tough, of course). Beyond that, other timing effects would be
avoidable, especially nothing would change on a quest level while you're
not on that level (the usual behaviour). And yes, to fight they should not
wait until they are adjacent. Aggravation effects would be appropriate;
their grade could be also depending on the ratio of living friends vs. foes.

>> c) Being quick to reach the quest leader should be worth something.

That would of course be contra-productive if the player shall engage in
the combat. (But I already mentioned that I largely dislike in this game
to play against a timer anyway. YMMV.)

>> [...]


>
>> d) The more defenders survive, the better.
>
> How would that be accomplished that the player gets an advantage?

See above for an example; easier travel back across the level. (And in the
unlikely(?) case that Quest leader happens to get killed then you won't get
your quest artifact identified.[*])

Janis

[*] Which would unfortunately make another "quality of text message" problem
apparent; the Quest leader, per default, removes the overly powerful powers
from the artifact when you bring it back. So those powers will be available
in case that the Quest leader dies? (That "problem" already exists in the
current version, BTW.)

David Ploog

unread,
Jun 7, 2009, 6:24:53 PM6/7/09
to
On Sun, 7 Jun 2009, Janis Papanagnou wrote:
> David Ploog wrote:

>> The proposal introduces a minigame; [...]
>
> Sure. I like the idea of minigames in Nethack, and another and new one is
> certainly worth discussing.
>
>> if you want to play it optimal, you would make sure that you bring useful
>> items when reaching the levels where the portal can occur (wands of secret
>> detection, magic mapping for the portal hunt; various items for the quest
>> leader hunt and still other items for killing the quest attacks in quick
>> ways).
>
> You imply (to solve the issue) existence of items that are not very common
> to have on levels around 11-16, IME.

These are just items with higher priority (if you want a headstart at the
minigame). There might be cases where you luck out -- I found it
interesting to note (when playtesting) how many turns you can save when
you think consciously about it. (Obviously, I never did speedruns before.)

> You should also consider that you need to be XL:14.

There might be a misunderstanding here. The minigame, as proposed by me,
is finished when you reach the Quest leader. It has nothing to do with
when you proceed deeper. The idea is that your leader has called for you
in despair; you showing up is good enough. Should you need more
experience, that cannot be helped. The leader can rest assured that you're
around.

> Immediately doing the Quest, once on the portal level,
> is untypical; usually I have to proceed much deeper (Castle/Valley) until
> I am admitted. Also; some quests are far too challenging for many character
> classes to be able to immediately enter them, you need to prepare for
> that.

I know. This is why my proposal does not demand any of this :)

>> (Note: I realise that the quests are not an urgent construction site for
>> Nethack. But I believe that the current system is unlogical, to the point
>> of being distorted, and could be improved in a rather simple manner.)
>
> Hmm.. - Have you, maybe, any idea how to improve the Quest logic without
> introducing timing effects? (Which would be the best way, IMO.)

Simply changing the text and messages, removing the idea of urgency.

> Or a way to improve your idea so that the random differences, which are
> typically huge between different instances of the game and between
> different classes played, wouldn't make the "rewards" (whatever that
> would be) arbitrary. I think answering those questions will make your
> "YANI" type of idea a more serious feature.

I don't expect my old idea to go in :)

You can make it much less random by skipping the first time check (the one
which counts how long you spent between getting the message and entering
the portal). This would allow you to prepare for the Home:1-only minigame
as long as you like.

>> I also believe that you (Janis) are among the truest of the true. If I
>> wanted to picture a man who is the last worshipper of a religion, it could
>> be you :)
>
> Interesting picture concerning someone who (not yet) believes in stories
> from any religion, including the earthly "religions" (opinions) we are
> told to believe every other day. ;-) But it is certainly correct that I
> don't follow every new proposal or worship the ideology that *any* change
> to the game is better than only sophisticated thought-out changes.

Fair enough.
Once I found myself in the designer seat, I immediately became much more
conservative as well. Even *good* ideas may be inappropriate for some
reason.

> And you should be really careful with such statements; they don't make a
> proposed idea better.

Sure. I am unattached enough to it, even if my postings here may prove the
opposite. In effect, I just wanted people here to keep from discussing
conducts at length when there's new content and design to be discussed. A
pretty selfish endavour, I guess I stop.

David

> Are there any ideas, meanwhile, what the price could be? If the feature
> will not be elaborated beyond the currently discussed state then I'd
> favour _Score Points_ as reward, lots of Score Points! You know why :-p

Yes, there have been two ideas (and even coded, though I had to start
quite some search in order to find the compueter):

1) The quest leader improves your gear (worn armour, wielded weapon).
Number of enchantment tries depends on your performance (how fast you've
been, how many attackers you killed, how many defenders got lost); it
should be something between 0 (bad performance) and 10 (exceptional
performance). Each enchantment target is used on a randomly chosen target,
one of your worn armour pieces and your wielded weapon(s). (I think we had
no special code for the case of a naked player arriving at the quest
leader.) Crucial point: the quest leader will not destroy any items.
Thus, you could try to use the quest minigame to get a good item (say
Mjollnir) beyond the safe enchantment limit. Of course, doing so would
mean that you forfeit a number of enchantments to be used otherwise. (E.g.
if you have ordinary armour and achieve a score of 8, you could expect an
AC gain of 7 or 8.) I really like the tradeoff involved here and it could
be used for something else instead. But having the quest leader do it
seems particularly interesting: it is a role already established and it
makes a lot of sense from the story told in the game. (Apart from the
Rogue quest, I should say.)

2) The quest leader creates another portal from Home:1, to a level in the
main dungeon shallower than the one with the standard quest portal. You
would gain a cycle in the dungeon graph this way (since you can go from
D:a to D:b either via the portals, or via the main dungeon). However, I
don't think that was a particularly good idea. I would not recommend to
use this prize anymore. (Ironically, there were some issues with the
portal which kept us from finishing the patch.)

I find score to be too metagamey a prize.

Cheers,
David

David Damerell

unread,
Jun 8, 2009, 11:02:05 AM6/8/09
to
Quoting Doug Freyburger <dfre...@yahoo.com>:
>I continuing playing 3.1.1 for about 2 years after 3.4.3 was
>released.

I'm kind of curious as to what 3.1.1 was a typo for.
--
David Damerell <dame...@chiark.greenend.org.uk> Oil is for sissies
Today is First Brieday, June.

Link

unread,
Jun 8, 2009, 11:07:10 AM6/8/09
to
On Jun 4, 11:08 am, David Ploog <pl...@mi.fu-berlin.de> wrote:
> On Mon, 1 Jun 2009, Patric Mueller wrote:
> > Announcing the first development release of UnNetHack.
>
> I find it very bothersome that this announcement did not trigger any
> feedback (apart from Derek of Spork and me, a renegade). As I see it,
> there is a simple explanation:
>
> a) The group is dead.
> b) Only the devteam fan club is still here.

I personally don't understand how this variant is significantly
different from normal Nethack. I have played SLASHEM and Sporkhack,
as their differences were very clear: SLASHEM added a giantic amount
of random stuff to the game, and Sporkhack wants to kill you dead.

I would be much more likely to comment on the gameplay if there were
detailed spoilers available. For example, UnNetHack claims to have
new monsters and new levels, but it doesn't really go into any detail
on the new stuff (e.g. look at http://www.spod-central.org/~psmith/nh/gazetteer/castle.html
for an example of a detailed level spoiler).

Doug Freyburger

unread,
Jun 8, 2009, 11:28:47 AM6/8/09
to
David Damerell <damer...@chiark.greenend.org.uk> wrote:

> Quoting  Doug Freyburger  <dfrey...@yahoo.com>:
>
> >I continuing playing 3.1.1 for about 2 years after 3.4.3 was
> >released.
>
> I'm kind of curious as to what 3.1.1 was a typo for.

3.3.1

Patric Mueller

unread,
Jun 8, 2009, 12:20:30 PM6/8/09
to
David Damerell <dame...@chiark.greenend.org.uk> wrote:
> Quoting Doug Freyburger <dfre...@yahoo.com>:
>>I continuing playing 3.1.1 for about 2 years after 3.4.3 was
>>released.
>
> I'm kind of curious as to what 3.1.1 was a typo for.

I would say: 3.3.1

3.4.1 wouldn't have been worth mentioning and anything older than
3.3.0 - which came out in December 1999 - would have been unbelievable
old.

David Ploog

unread,
Jun 8, 2009, 12:36:02 PM6/8/09
to
On Mon, 8 Jun 2009, Link wrote:
> On Jun 4, 11:08�am, David Ploog <pl...@mi.fu-berlin.de> wrote:
>> On Mon, 1 Jun 2009, Patric Mueller wrote:
>>> Announcing the first development release of UnNetHack.
>>
>> I find it very bothersome that this announcement did not trigger any
>> feedback (apart from Derek of Spork and me, a renegade). As I see it,
>> there is a simple explanation:
>>
>> a) The group is dead.
>> b) Only the devteam fan club is still here.
>
> I personally don't understand how this variant is significantly
> different from normal Nethack.

Unihorns, wishes, genocide? Heck2 and other changes for Gehennom? From
what I recall about the typical Nethack discussion, some of the most
pressing issues have been addressed.

I am not saying that Un does it perfect. But no response in either
direction simply proves that people don't care. Which is interesting
because the annual another-year-without-release debate always ends with
"less talk, more action".

David

Patric Mueller

unread,
Jun 8, 2009, 3:02:36 PM6/8/09
to
Link <chill...@yahoo.com> wrote:
>
> I personally don't understand how this variant is significantly
> different from normal Nethack.

At the current state of development it depends how you're playing and
what you like.

If you're a notorious quaffer of fountains and blessed smoky potions
and you wish for your favorite DSM at the first possible moment then
it is significantly different.

If you're bored by the same levels over and over again UnNetHack would
be worth a try.

> I have played SLASHEM and Sporkhack,
> as their differences were very clear: SLASHEM added a giantic amount
> of random stuff to the game, and Sporkhack wants to kill you dead.

I disagree that SporkHack wants to kill you dead. That's also
Slash'Em's task.

SporkHack's stated purpose is among others to "try to make the game
more interesting for experienced/skilled players".

But "more interesting" in NetHack often means "more often dead".

> I would be much more likely to comment on the gameplay if there were
> detailed spoilers available. For example, UnNetHack claims to have
> new monsters and new levels,

More or less everything that differs in UnNetHack has been described
on the project with a note where I got it from if it's not original to
UnNetHack.

A lot of the stuff in UnNetHack has either been ripped out of
variants, like the Blackmarket from Slash'Em or the Orcish Mine Town
from SporkHack.

There are also many patches integrated into UnNetHack that have been
discussed in this group.

Most of L's stuff has been discussed in this group, so a google groups
search should turn something up.

> but it doesn't really go into any detail
> on the new stuff (e.g. look at http://www.spod-central.org/~psmith/nh/gazetteer/castle.html
> for an example of a detailed level spoiler).

Good spoilers need time and work (and somebody who does them).

Look at Slash'Em for example. Before the arrival of WikiHack there
weren't detailed level description around.

And it looks like Pasi Kallinen's Medusa levels (which are also in
UnNetHack) didn't get an entry on WikiHack.

Janis Papanagnou

unread,
Jun 8, 2009, 4:50:59 PM6/8/09
to
David Ploog wrote:
> On Mon, 8 Jun 2009, Link wrote:
>>
>> I personally don't understand how this variant is significantly
>> different from normal Nethack.
>
> Unihorns, wishes, genocide? Heck2 and other changes for Gehennom? From
> what I recall about the typical Nethack discussion, some of the most
> pressing issues have been addressed.

The difference to your view seems to be the subjective line where the
difference between vanilla and some variant become "significant" enough
to deviate from the, errm.., advantage[*] of staying with vanilla. It's
just that simple.

> I am not saying that Un does it perfect.

While being perfect is an idealistic goal I think no one, in practise,
demands a perfect variant.

> But no response in either
> direction simply proves that people don't care.

Not that much as you seem to like them care. Yes, you're right.

> Which is interesting

Not really...

> because the annual another-year-without-release debate always ends with
> "less talk, more action".

Those (mostly boring) discussions had another primary claim; the proposer
demanded to inherit the vanilla branch without creating a variant himself.
Creating a variant does not automagically mean that it is considered worth
a switch because it fixes some issues.

I am well aware that Spork (which I playtested) and Unnethack (which I
haven't played yet) fix some issues. But so what?

People who are enthusiastic about those changes will probably play those
variants. But if you ask for discussion of the variants I suppose most has
already been discussed before in contexts of the involved patches.

Janis

[*] Which is certainly diminishing with every new year that passes without
new version.

David Ploog

unread,
Jun 8, 2009, 5:10:28 PM6/8/09
to
On Mon, 8 Jun 2009, Janis Papanagnou wrote:
> David Ploog wrote:
>> On Mon, 8 Jun 2009, Link wrote:
>>>
>>> I personally don't understand how this variant is significantly
>>> different from normal Nethack.
>>
>> Unihorns, wishes, genocide? Heck2 and other changes for Gehennom? From what
>> I recall about the typical Nethack discussion, some of the most pressing
>> issues have been addressed.
>
> The difference to your view seems to be the subjective line where the
> difference between vanilla and some variant become "significant" enough
> to deviate from the, errm.., advantage[*] of staying with vanilla. It's
> just that simple.

Yes. I think I only got so much worked up because r.g.r.n was semi-lively
with that conduct discussion. And as fas I am concerned, conducts have
been introduced because the game is too easy. So instead of discussing how
to make it interesting (again), people stick to the shifted goalpost.

> I am well aware that Spork (which I playtested) and Unnethack (which I
> haven't played yet) fix some issues. But so what?
>

> [*] Which is certainly diminishing with every new year that passes without
> new version.

I really understand both points. But these two together could imply that
the Devteam's policy is making Nethack's own grave. (In a stronger form as
you'd have for a project that was simply abandoned in silence: for
Nethack, the ominous presence of vanilla and the devteam hurts all forks.)

David

Patric Mueller

unread,
Jun 8, 2009, 6:01:41 PM6/8/09
to
David Ploog <pl...@mi.fu-berlin.de> wrote:
> On Sun, 7 Jun 2009, Janis Papanagnou wrote:
>> David Ploog wrote:
>
>> And you should be really careful with such statements; they don't make a
>> proposed idea better.
>
> Sure. I am unattached enough to it, even if my postings here may prove
> the opposite. In effect, I just wanted people here to keep from
> discussing conducts at length when there's new content and design to
> be discussed. A pretty selfish endavour, I guess I stop.

Discussing conducts is easy because everybody can judge how a specific
conduct would affect their normal game play.

For a new variant this is much harder as the players just don't have
the same deep understanding of the game mechanics as with vanilla
NetHack.

> 1) The quest leader improves your gear (worn armour, wielded weapon).
> Number of enchantment tries depends on your performance (how fast
> you've been, how many attackers you killed, how many defenders got
> lost); it should be something between 0 (bad performance) and 10
> (exceptional performance). Each enchantment target is used on a
> randomly chosen target, one of your worn armour pieces and your
> wielded weapon(s). (I think we had no special code for the case of a
> naked player arriving at the quest leader.) Crucial point: the quest
> leader will not destroy any items.

I like that. This would open up a possibility to let the player create
a personal unique item. This leads to more emotional involvement by
the player. At least I get irrationaly attached to such things in
computer games. :)

> I find score to be too metagamey a prize.

And in Vanilla not really worth the hassle. Score is easily
accumulated. In UnNetHack OTOH the scoring is different enough to have
more meaning.

David Ploog

unread,
Jun 8, 2009, 10:09:40 PM6/8/09
to
On Tue, 9 Jun 2009, Patric Mueller wrote:
> David Ploog <pl...@mi.fu-berlin.de> wrote:

>> 1) The quest leader improves your gear (worn armour, wielded weapon).
>> Number of enchantment tries depends on your performance (how fast
>> you've been, how many attackers you killed, how many defenders got
>> lost); it should be something between 0 (bad performance) and 10
>> (exceptional performance). Each enchantment target is used on a
>> randomly chosen target, one of your worn armour pieces and your
>> wielded weapon(s). (I think we had no special code for the case of a
>> naked player arriving at the quest leader.) Crucial point: the quest
>> leader will not destroy any items.
>
> I like that. This would open up a possibility to let the player create
> a personal unique item. This leads to more emotional involvement by
> the player. At least I get irrationaly attached to such things in
> computer games. :)

I liked it too, obviously. Especially as in almost all cases, it would be
more sensible to spread the enchantments across as many armour pieces and
weapons as possible. Us fools would still try for that +9 item every time
I guess :)

>> I find score to be too metagamey a prize.
>
> And in Vanilla not really worth the hassle. Score is easily
> accumulated. In UnNetHack OTOH the scoring is different enough to have
> more meaning.

But even so: score is much less rewarding than something visible in-game.

David

JoaoSantos

unread,
Jun 9, 2009, 8:13:55 AM6/9/09
to
On Jun 5, 9:55 pm, Patric Mueller <bh...@bigfoot.com> wrote:
[...]
> There's no public server yet.
>
> But it is planned.
>
> As soon as it goes live, it also will be announced here.
>
> Bye
> Patric


I completely lost interest on local games, so I only play on servers.
When there is a server for UnNethack I'll give it a try, like I did
for other variants.
Without looking down on the player base and amount of effort spent on
Spork, I wonder what are the requirements for a nethack variant to be
playable on nethack.alt.org

solidsnail

Link

unread,
Jun 9, 2009, 10:08:00 AM6/9/09
to
On Jun 8, 3:02 pm, Patric Mueller <bh...@bigfoot.com> wrote:

> Link <chillyn...@yahoo.com> wrote:
> > I have played SLASHEM and Sporkhack,
> > as their differences were very clear: SLASHEM added a giantic amount
> > of random stuff to the game, and Sporkhack wants to kill you dead.
>
> I disagree that SporkHack wants to kill you dead. That's also
> Slash'Em's task.

SLASHEM does not want me dead. SLASHEM does have some race/class
combinations that are harder than the average Nethack race/class
combinations, but it also has race/class combinations that are far
easier than anything that is in Nethack.

Example 1: Doppleganger Monk. If anything remotely dangerous appears,
turn into a black dragon, and blast it with your disintegration
attack. If you need magic resistance, reflection, drain resistance,
or stoning resistance -- then you can turn into whatever monster has
that. And since you get warning at XL7, you can know what you need to
turn into, before the monsters get near you.

Example 2: Gnomish Healer. They can do "Tinker farming", which I
think it even more broken than pudding farming, since it is way way
faster to tinker farm. Since they start with the stone to flesh spell,
they can get "infinite" food from boulders, which gives them more than
enough time to tinker farm. When they hit XL7, they can easily take
common items, and turn them into a cloak of magic resistance, helm of
telepathy, gauntlets of swimming, and a magic whistle. The first sack
they get can turn into a bag of holding. Any crystal balls they get
can be turned into magic markers. Any flint stones they get can be
turned into healthstones. The first leash they get can be turned into
a saddle. Any leather armor (or leather jacket) they find can be
turned into studded leather armor (which is good enough for AC, since
the healer can easily get 9 points of protection).

The end result is that once they hit XL7, they can easily have -5 AC
(or lower), and a stupid amount of powerful items. And they can use
healing spells. And the healing spells + stethoscope + magic whistle
makes it easy to manage pets, which will make it easy to survive until
your character is powerful enough to run around the dungeon by himself.

Simon

unread,
Jun 10, 2009, 5:25:24 AM6/10/09
to
On Jun 1, 5:46 pm, Patric Mueller <bh...@bigfoot.com> wrote:
> Announcing the first development release of UnNetHack.
>
> The main intent of this fork is to put more randomness, challenges and
> fun into NetHack.
>
> It features more levels (e.g. all of Pasi Kallinens new levels, most
> of the Heck² levels by Kelly Bailey and the Blackmarket from
> Slash'Em), several UI changes like automatically opened doors and the
> pickup thrown patch and a lot of game play related changes.
>
> All changes are documented on the trac project page.
>
> On irc.freenode.net there's a channel #unnethack for discussion.
>
> Trac project page:http://apps.sourceforge.net/trac/unnethack-12312/
>
> Source code and Win32 tty download:http://sourceforge.net/project/showfiles.php?group_id=250747&package_...
>
> Some screenshots:http://apps.sourceforge.net/trac/unnethack-12312/wiki/Screenshots

This looks excellent - duly downloaded and will start playing it
soon. Not a radical departure, more an evolution of vanilla to
incorporate some of the changes that the devteam should have added
over the last five and a half years.

I tend to play locally, on a laptop, when travelling. Does UNH
currently support Hearse? And if not are there any intentions to
support it in the future?

Any thoughts on the biodiversity patch?

Keep up the good work - this is precisely what is needed!

S

Patric Mueller

unread,
Jun 10, 2009, 1:21:55 PM6/10/09
to
Simon <yarog...@googlemail.com> wrote:
>
> This looks excellent - duly downloaded and will start playing it
> soon. Not a radical departure, more an evolution of vanilla to
> incorporate some of the changes that the devteam should have added
> over the last five and a half years.

That was more or less how I worked first week of development.

I almost solely added patches to see how far one can quickly deviate
from Vanilla with those patches that have been floating around for
ages.

Those patches get you very far in terms of look (like menucolors, new
levels) but there are fewer patches that change game mechanics (like
Chi's Vlad patch).

> I tend to play locally, on a laptop, when travelling. Does UNH
> currently support Hearse? And if not are there any intentions to
> support it in the future?

In principle it should work. There have been no changes to nethacks
save/bones file format.

# ./bones-info -v bonD0.6
bonD0.6: 17991 bytes
read as: little endian
incarnation: 50659584 (3.5.1.0)
feature_set: 4066510 (REINCARNATION SINKS UNKNOWN-3 KOPS MAIL TOURIST
STEED TEXTCOLOR INSURANCE ELBERETH EXP_ON_BOTL
SCORE_ON_BOTL)
entity_count: 572617096 (34 artifacts, 535 objects, 392 monsters)
struct_sizes: 2827064780 (168 flag, 64 obj, 101 monst, 460 you)

But I don't know how bones of an unknown program version would be
treated by the hearse server.

If I look at http://hearse.krollmark.com/bones.dll?act=versionstats
it seems that Hearse doesn't yet support SporkHack?

> Any thoughts on the biodiversity patch?

Currently only the disintegrator has been snatched from the
biodiversity patch.

Some of the more interesting monsters like the watersprout gargoyle
are likely candidates to make it into a future release.

The biodiversity patch is huge. I think I haven't yet looked at
everything it changes. So if you want to draw my attention to
something specific, just say it.

> Keep up the good work - this is precisely what is needed!

Thanks!