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 Historical content remains viewable.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

[evilhack] Game terminating surprises - and solutions?

121 views
Skip to first unread message

Janis Papanagnou

unread,
Mar 26, 2023, 1:18:16 PM3/26/23
to
It's certainly interesting to know and to collect experiences about
what to expect, and more importantly, how to face or handle game
terminating surprises; and especially early game. Given some recent
advertisement about "a learning curve" in EvilHack; what is it that
we can actually learn from all that...?

Personally it seems to me that - with arbitrary changes at least -
that can only be anticipated by being fully spoiled; this basically
means source code studies, since the (indeed long but in that respect
not very helpful) article in Wikipedia doesn't the least cover the
basics. This is what I have learned (on the meta-level).

Some recent cases; yesterday I wrote that
Recently I got an arch-lich at the orc-haunted mine-town variant.
Okay, bad luck! - Hmm.., bad luck, or bad game design? - Nethack
has occasionally also shape changers that appear in deadly shapes
early game. (Have the odds changed in this variant? Subjectively
it seems to me that this unbalance has become standard here.)

I already reported about unexpected and _arbitrary_ attack types,
like deadly illness from quite early appearing locusts; which has
no counterpart in real life (and I wondered about any rationale
for that). - What to do? - recently I died the second time (this
time in the Rat's mines-end level) to locusts hordes; no escape
path, no hilarious stack of any healing potions this time, that
would have been necessary.

And just now I died from an adult purple worm. - Where and When? -
After only a few minutes playing, a few turns only, and soon after
entering dungeon level 3. - What can I learn from that; dungeon
levels below level 2 should not be entered without a wand of
death? (Which wouldn't have helped in a 1-turn swallow-and-die
attack.)

Janis

Pat Rankin

unread,
Mar 26, 2023, 4:12:53 PM3/26/23
to
On Sunday, March 26, 2023 at 10:18:16 AM UTC-7, Janis Papanagnou wrote:
[...]
> And just now I died from an adult purple worm. - Where and When? -
> After only a few minutes playing, a few turns only, and soon after
> entering dungeon level 3. - What can I learn from that; dungeon
> levels below level 2 should not be entered without a wand of
> death? (Which wouldn't have helped in a 1-turn swallow-and-die
> attack.)

This could happen in nethack. Each time a shrieker shrieks,
there is a chance that a purple worm will be created on the
level. (The location will be random rather than forced to be
nearby.) When a purple worm swallows you, the number of
turns it takes to be digested is based on your AC and
constitution. The initial bite inflicts some damage so total
digestion will have a head start.

Chance per shriek that a monster will be created is 1 in 10
and when that occurs, 1 in 13 that it will be a purple worm,
remaining 12 out of 13 something random appropriate for
the current level. To-be-3.7 has added a change to make a
baby purple worm instead of an adult one when the latter is
considered to be too difficult. Evilhack might have left that
out, or perhaps has a looser notion of "too difficult", or both.

Janis Papanagnou

unread,
Mar 27, 2023, 12:57:28 AM3/27/23
to
On 26.03.2023 22:12, Pat Rankin wrote:
> On Sunday, March 26, 2023 at 10:18:16 AM UTC-7, Janis Papanagnou wrote:
> [...]
>> And just now I died from an adult purple worm. - Where and When? -
>> After only a few minutes playing, a few turns only, and soon after
>> entering dungeon level 3. - What can I learn from that; dungeon
>> levels below level 2 should not be entered without a wand of
>> death? (Which wouldn't have helped in a 1-turn swallow-and-die
>> attack.)
>
> This could happen in nethack. Each time a shrieker shrieks,

Do shriekers appear that early? (Dlvl:3, XL:~1) - Anyway; there
was no shrieker around. I experienced shriekers to shriek only
when adjacent in combat [in Nethack]. (Has that changed, or do
I misremember?)

In EvilHack there's also the possibility that monsters fight
each other, IIUC, thus possible shriekers might shriek without
involving the player. (I don't recall that I've heard anything
like that, though.)

My first thought was a shape-changer (a chameleon), but would
that appear that early? Anyway, the result is the same; whether
in Nethack or any other variant, it's unbalanced (in my book).

Janis

> [...]


Janis Papanagnou

unread,
Mar 27, 2023, 1:05:37 AM3/27/23
to
On 26.03.2023 19:18, Janis Papanagnou wrote:
>
> [...] - What can I learn from that; dungeon
> levels below level 2 should not be entered without a wand of
> death? [...]

Current game, a narrow early shop with mimics. Wanted to lure
one mimic out of the shop; sadly it was (that early) a large
mimics and I could not escape. Without Elbereth I took the risk
to zap an unidentified wand at it - and there it was, the above
mentioned ("lucky") wand of death! -, but unfortunately it was
cursed, it thus "backfired" (a EvilHack feature?) and outright
killed me. (No, I don't blame the game for that. It's actually
quite funny! But it fits the pattern.)

Janis

Keith Simpson

unread,
Mar 27, 2023, 5:23:33 PM3/27/23
to
In EvilHack, a cursed wand has a small chance to backfire and target you instead. This is also present in xNetHack, Hack'EM, and possibly a couple other variants. Sometimes hard to keep track of what all exists in which variant.

Keith Simpson

unread,
Mar 27, 2023, 5:25:26 PM3/27/23
to
How shriekers and baby/adult purple worms behave in EvilHack is no different than vanilla NetHack.

Keith Simpson

unread,
Mar 27, 2023, 5:38:15 PM3/27/23
to
On Sunday, March 26, 2023 at 1:18:16 PM UTC-4, Janis Papanagnou wrote:
> Personally it seems to me that - with arbitrary changes at least -
> that can only be anticipated by being fully spoiled; this basically
> means source code studies, since the (indeed long but in that respect
> not very helpful) article in Wikipedia doesn't the least cover the
> basics. This is what I have learned (on the meta-level).

When Hack and NetHack first came out, there were no spoilers, at least none that I knew about. The 'world wide web' didn't exist as it does today back then. And not everybody knows how to source dive - I certainly didn't when I first started playing. I didn't even know about Elbereth until after my second ascension. So you play, you experience what comes your way, you remember/take notes, and if you die, you try again, armed with the new knowledge you've acquired. That's how it was done 'back in the day' and frankly it's a bit distressing to see some players who won't try out *any* new game unless they're completely spoiled beforehand. You can spoil yourself if you want to, the wiki has a rather large entry for EvilHack, but I would recommend experiencing it fresh. As long as you've been playing, wouldn't that be refreshing in a way?

> Some recent cases; yesterday I wrote that
> Recently I got an arch-lich at the orc-haunted mine-town variant.
> Okay, bad luck! - Hmm.., bad luck, or bad game design? - Nethack
> has occasionally also shape changers that appear in deadly shapes
> early game. (Have the odds changed in this variant? Subjectively
> it seems to me that this unbalance has become standard here.)

Alex, I'll take 'bad luck' for $800 please.

Shape changer odds of spawning hasn't changed from regular NetHack, neither have poly trap odds in the mines. I'm betting something stepped on a poly trap and you just got an unlucky RNG roll. It happens. I'm confident it's not due to bad game design.

> I already reported about unexpected and _arbitrary_ attack types,
> like deadly illness from quite early appearing locusts; which has
> no counterpart in real life (and I wondered about any rationale
> for that). - What to do? - recently I died the second time (this
> time in the Rat's mines-end level) to locusts hordes; no escape
> path, no hilarious stack of any healing potions this time, that
> would have been necessary.
>
> Janis

Locusts and their attack type was imported from SporkHack. That variant has been around for a *long* time - given your experience, I'm surprised you hadn't ever come across it until now.

RecRanger

unread,
Mar 27, 2023, 5:57:53 PM3/27/23
to

> > Some recent cases; yesterday I wrote that
> > Recently I got an arch-lich at the orc-haunted mine-town variant.
> > Okay, bad luck! - Hmm.., bad luck, or bad game design? - Nethack
> > has occasionally also shape changers that appear in deadly shapes
> > early game. (Have the odds changed in this variant? Subjectively
> > it seems to me that this unbalance has become standard here.)
> Alex, I'll take 'bad luck' for $800 please.
>
> Shape changer odds of spawning hasn't changed from regular NetHack, neither have poly trap odds in the mines. I'm betting something stepped on a poly trap and you just got an unlucky RNG roll. It happens. I'm confident it's not due to bad game design.

Minetown cannot generate at level 8, though, can it? Unless vanilla or
Evil changes something? Poly traps do not generate until level 8.


--

RecRanger

unread,
Mar 27, 2023, 6:02:37 PM3/27/23
to
Okay, I am seeing now that it can happen. Was this a revision in 3.6.x?

--




Janis Papanagnou

unread,
Mar 27, 2023, 9:09:24 PM3/27/23
to
On 27.03.2023 23:38, Keith Simpson wrote:
> On Sunday, March 26, 2023 at 1:18:16 PM UTC-4, Janis Papanagnou
> wrote:
>> Personally it seems to me that - with arbitrary changes at least -
>> that can only be anticipated by being fully spoiled; this
>> basically means source code studies, since the (indeed long but in
>> that respect not very helpful) article in Wikipedia doesn't the
>> least cover the basics. This is what I have learned (on the
>> meta-level).
>
> When Hack and NetHack first came out, there were no spoilers, at
> least none that I knew about.

There have always been spoilers. (That you don't know them should
make you a bit more reserved, also with your imputations below.)

I haven't played Nethack when its first release came out. I started
with NH-3.0.9 (in the early 1990's, IIRC). I played it continuously
until NH-3.4.3, until NAO stopped its support for it.

I tried some NH-3.6.x version, played (I think) two games, ascended
one of it. Didn't like it. Quite some changes have obviously been
promoted by the Spork developer who became member of the post NH343
development team, and his footprint can certainly be seen.

> The 'world wide web' didn't exist as it does today back then.

The WWW was unnecessary since Usenet was much more populated at
these times, specifically this newsgroup. You could get so much
information and exchange experiences that you could become spoiled
to any degree you wanted.

Today this newsgroup is effectively dead. (Nobody [here] seems to
care about Nethack any more, and if I'd not occasionally have had
populated it with [slashem] posts or lately (sort of) spammed it
with [evilhack] posts...) So this previously important channel
effectively drained as source for information exchange, spoilers,
etc.

The NH-Guidebook was good enough to get a well informed start. With
that and with some caution based on common sense players were able
to reach far even if unspoiled. (If own experiences don't match or
if memories are faint have a look at the "Ellora" story as a hint.)

Of course there are single special cases that must be experienced
the hard way (like an exploding bag of holding[*]), but a lot other
things could be derived (like Medusa turning one to stone), or that
it would be - here in EvilHack[**] - unwise to wear "the One Ring"
as it comes cursed.

[*] Actually an unnecessary nuisance, that could be fixed in a more
balanced way (i.e. without the guaranteed frustration of losing
quite _everything_ carried that was collected during a long game);
and there's many options I can think of here to punish the player
(for his ignorance or sloppy play) without completely frustrating
him/her.

[**] Yes, this is a positive example that something is derivable.

> And not everybody knows how to source dive - I
> certainly didn't when I first started playing. I didn't even know
> about Elbereth until after my second ascension. So you play, you
> experience what comes your way, you remember/take notes, and if you
> die, you try again, armed with the new knowledge you've acquired.
> That's how it was done 'back in the day' and frankly it's a bit
> distressing to see some players who won't try out *any* new game
> unless they're completely spoiled beforehand. You can spoil yourself
> if you want to, the wiki has a rather large entry for EvilHack, but I
> would recommend experiencing it fresh. As long as you've been
> playing, wouldn't that be refreshing in a way?

This is actually what I am (or what I was) doing with EvilHack.
(Should be quite obvious if you inspect the many posts/advertisment
for [evilhack] I've posted recently.

In the past I've tried out several roguelike variants (omega, ularn,
moria, adom), but none (IMO) competed with Nethack. Currently I am
also trying out several variants (as you see), but from those I have
seen only Slashem appears appealing [to me] (despite a _few_ things
that I dislike); Slashem made quite some excellent design decisions.

But 90+% unexpected, underivable effects in a game... - how could that
be "refreshing". (That's at best frustrating, because common sense
doesn't help.) And the whole balance factor is, erm.., disputable in
that variant (as far as my early/midgame experience and my observations
of other games on a server goes).

(BTW, I wasn't attentive; are you maybe the developer of EvilHack?
that would - with the developer knowledge - at least explain why you
are so keenly defending it. Being open to criticism might help. If
you don't care, I care even less, mind.)

>
>> Some recent cases; yesterday I wrote that Recently I got an
>> arch-lich at the orc-haunted mine-town variant. Okay, bad luck! -
>> Hmm.., bad luck, or bad game design? - Nethack has occasionally
>> also shape changers that appear in deadly shapes early game. (Have
>> the odds changed in this variant? Subjectively it seems to me that
>> this unbalance has become standard here.)
>
> Alex, I'll take 'bad luck' for $800 please.
>
> Shape changer odds of spawning hasn't changed from regular NetHack,
> neither have poly trap odds in the mines. I'm betting something
> stepped on a poly trap and you just got an unlucky RNG roll. It
> happens. I'm confident it's not due to bad game design.

The RNG decisions are no bad game design, sure.

But bad (unbalanced) design is, say, if polytraps appear at too
shallow levels (arch-lich at minetown?), or if the outcome of a shape
transformation is far out of depth. Okay, you may say; "shit happens".
But such design consequences are not "refreshing" but just foster
frustration.

>
>> I already reported about unexpected and _arbitrary_ attack types,
>> like deadly illness from quite early appearing locusts; which has
>> no counterpart in real life (and I wondered about any rationale for
>> that). - What to do? - recently I died the second time (this time
>> in the Rat's mines-end level) to locusts hordes; no escape path, no
>> hilarious stack of any healing potions this time, that would have
>> been necessary.
>
> Locusts and their attack type was imported from SporkHack. That
> variant has been around for a *long* time - given your experience,
> I'm surprised you hadn't ever come across it until now.

Frankly, I cannot say whether I've played/ascended a Spork game or two
or not, I really just don't recall. But above I've already said what I
think about many of the design changes in Spork (that sadly found also
their way into Nethack, and as you say obviously also in EvilHack).

In any way; implementing (or borrowing from another game) some stupid
(=unbalanced, =non-derivable) "locust" concept doesn't make it any
better. YMMV.

Janis
--
BTW, how brain-dead is it to change the "satiated / continue eating"
default value from 'no' to 'yes'. - Variant developers should take at
least minimum care to enhance (not worsen) the user interface of their
games. (And that's not the only UI issue some newer roguelikes have.)

I hate bad user interfaces. And implementations seem to degrade instead
of learning from sophisticated implementations (or from common sense,
unless not available).

Janis Papanagnou

unread,
Mar 27, 2023, 9:16:47 PM3/27/23
to
On 27.03.2023 23:23, Keith Simpson wrote:
> On Monday, March 27, 2023 at 1:05:37 AM UTC-4, Janis Papanagnou
> wrote:
>> On 26.03.2023 19:18, Janis Papanagnou wrote:
>>>
>>> [...] - What can I learn from that; dungeon levels below level 2
>>> should not be entered without a wand of death? [...]
>>
>> Current game, a narrow early shop with mimics. Wanted to lure one
>> mimic out of the shop; sadly it was (that early) a large mimics and
>> I could not escape. Without Elbereth I took the risk to zap an
>> unidentified wand at it - and there it was, the above mentioned
>> ("lucky") wand of death! -, but unfortunately it was cursed, it
>> thus "backfired" (a EvilHack feature?) and outright killed me. (No,
>> I don't blame the game for that. It's actually quite funny! But it
>> fits the pattern.)
>
> In EvilHack, a cursed wand has a small chance to backfire and target
> you instead. This is also present in xNetHack, Hack'EM, and possibly
> a couple other variants. Sometimes hard to keep track of what all
> exists in which variant.

Don't expect EvilHack players to know all (or the mentioned) variants.

But mind that I didn't complain; it is completely reasonable to expect
bad things from unidentified wands. My first thought in that hopeless
situation was that it might be an ineffective wand (probing, nothing,
light, etc.) or even an attack wand (fire came to my mind, since I was
in a scrolls shop). But it didn't occur to me that it could be death,
a rare wand. That an unidentified wand may be cursed with arbitrary
effects, like exploding, or some new, like backfiring, lies within the
range of the expectable behaviors. At that stage it was a desperate
move since I'd have died anyway, one way or another.

Janis

Pat Rankin

unread,
Mar 28, 2023, 3:47:03 PM3/28/23
to
On Monday, March 27, 2023 at 6:09:24 PM UTC-7, Janis Papanagnou wrote:
[...]
> BTW, how brain-dead is it to change the "satiated / continue eating"
> default value from 'no' to 'yes'. [...]

This is out of line. The old prompt was "stop eating?" with <space>
as a synonym for 'no' and there were valid complaints about that
over the years. The "new" prompt is "continue eating?" and <space>
is still 'no' but it now picks the safer choice. 'y' and 'n' are reversed
from what they were in versions released twenty or more years ago
but If you decide to answer prompts without reading them, that's
on you, not on the game's design.

You started playing back when there was no '{stop|continue} eating?'
prompt at all and I don't recall ever seeing you complain about a
prompt being introduced for it. There used to be a compile-time
option called HARD which was enabled by default. Choking had no
prompt at all under that. If the program was custom built with it
disabled, eating which could cause risk of choking would give a
prompt before the first bite (using the old backwards 'stop?'
phrasing). When HARD was eliminated to cut out all the extra
testing and tuning it necessitated, most of the 'hard' features
remained as hard-coded instead of optional. However, the
current scheme of prompting if you survive the first bite got
introduced (inheriting the backward phrasing). The difference
from defunct not-HARD for it is that the check for whether to
issue a prompt comes after the first bite, so you might choke
without prompting if the first bite puts your satiation past an
internal threshold.

If Evilhack is too hard, nethack 3.6.x has undergone many
changes since the 3.6.0 you occasionally still complain about,
some to address things in 3.6.0 that players didn't like. The one
that comes immediately to mind is monster movement randomly
giving monsters an extra move. (For ones who get a fractional
amount of movement either above or below normal one move
per turn, the time when they'll get their fractional move has
stayed unpredictable compared with 3.4.3 and earlier, but it
no longer gets boosted like 3.6.0 had a small chance of doing.)

RecRanger

unread,
Mar 28, 2023, 5:19:54 PM3/28/23
to
On Tuesday, March 28, 2023 at 3:47:03 PM UTC-4, Pat Rankin wrote:
> On Monday, March 27, 2023 at 6:09:24 PM UTC-7, Janis Papanagnou wrote:
> [...]
> > BTW, how brain-dead is it to change the "satiated / continue eating"
> > default value from 'no' to 'yes'. [...]
>
> This is out of line. The old prompt was "stop eating?" with <space>
> as a synonym for 'no' and there were valid complaints about that
> over the years. The "new" prompt is "continue eating?" and <space>
> is still 'no' but it now picks the safer choice. 'y' and 'n' are reversed
> from what they were in versions released twenty or more years ago
> but If you decide to answer prompts without reading them, that's
> on you, not on the game's design.
>

Do not really recall how it happened, but I lost a tank of a character, that
was on their way to ascension, to this dumb change.

Choked.


--

Janis Papanagnou

unread,
Mar 29, 2023, 6:20:11 AM3/29/23
to
On 28.03.2023 21:47, Pat Rankin wrote:
> On Monday, March 27, 2023 at 6:09:24 PM UTC-7, Janis Papanagnou wrote:
> [...]
>> BTW, how brain-dead is it to change the "satiated / continue eating"
>> default value from 'no' to 'yes'. [...]
>
> This is out of line. The old prompt was "stop eating?" with <space>
> as a synonym for 'no'

I'm sorry that you missed the informal nature of "satiated / continue
eating" quote; the NH343 prompt was different compared to the EvilHack
prompt, and I wanted to keep it terse and wouldn't have thought that
anyone would miss (or ignore) the actual statement and fact behind it.
So as a service for you...
The _default_ of former versions of Nethack (I am speaking about 343)
was that you would _not_ continue with the fatal effect of choking to
death. The EvilHack defaults behave differently, exact the opposite;
you will die if you continue - and it's irrelevant why you did that
(by accident or by being used to a safer interface). Defaults are in
some cases even better to avoid completely, but if you have a default
it should never foster the critical process continuation; in real
life, in systems, in games, in software.
Nowadays you find sensible defaults everywhere; "Delete all Emails
from this folder?" - who would default that to "yes"?!

> and there were valid complaints about that over the years.

Are you speaking about newer Nethack version or about EvilHack? I'm
referring to EvilHack (as the subject indicates).

(What would be valid complaints against a behavior that the default
won't kill you by accident?)

> The "new" prompt is "continue eating?" and <space>
> is still 'no' but it now picks the safer choice. 'y' and 'n' are reversed
> from what they were in versions released twenty or more years ago
> but If you decide to answer prompts without reading them, that's
> on you, not on the game's design.

That's utter nonsense. It's not about reading it's about a sensible
user interface! If default values lead to deaths that's bad design.
(More so if that changed from version to version, from safe design
to unsafe design.)
(And I thought you'd have some CS background and would know at least
a bit about sensible UI design, if only from own personal experience.
After all you were working in the IT area since DEC VMS times, IIUC.
But not having that expertise AND posting nonsensical accusations
and lowbrow statements is certainly not what I've expected from you.)

>
> You started playing back when there was no '{stop|continue} eating?'
> prompt at all and I don't recall ever seeing you complain about a
> prompt being introduced for it. [...]

You recall about all my complaints and posts from past 30 decades?
I don't believe you; you must be a liar!

I recall that over the decades I had always "complained" about bad
user interface designs or bad features, usually by _suggestions_ of
better alternatives; I also recall that some suggestions had even
been incorporated. (The last one I recall - already very long ago -
was a bad menu design change in newer versions, and they then allowed
in the NAO version that the alternative form would still be accepted.
Another one, even longer back, was that when #naming you could not
remove a name, so a single blank has been defined to clear it. But
meanwhile that interface changed again - in a (IMO) bad way -, now
(in EvilHack, but maybe also in NH36x?) you have to delete all the
characters and then enter the space. - Just for a few old examples
that I still recall and some prospect.

For actual versions there's still a lot to complain about; e.g. the
inconsistent menu handling, or the paging. (Currently speaking about
EvilHack, but probably it's directly or indirectly borrowed from
Nethack.) But for an "oldie" like me it's not of relevance any more.
If people responsible for the design produce bullshit, hey, just let
them do it, why should I care. (You certainly noticed that the only
part of my previous post that you replied to was the one that I put
detached in the signature block. Peculiarly. Any now we have a huge
reply from you and an ever larger from me; for no purpose.)

If UI was/is (in parts or completely) bad and suggestions got/get
incorporated there's nothing to complain. That's a Good Thing.

On the user's side there's typically some expectation that systems
have a consistent sensible concept, and if changed get better, not
worse. (From experience we must admit that this is not always the
case.)

>
> If Evilhack is too hard, nethack 3.6.x has undergone many

(Difficulty and balance are completely different things. Slashem,
for example, is also "hard" but obviously a lot better balanced.)

> changes since the 3.6.0 you occasionally still complain about,
> some to address things in 3.6.0 that players didn't like. The one
> that comes immediately to mind is monster movement randomly
> giving monsters an extra move. (For ones who get a fractional
> amount of movement either above or below normal one move
> per turn, the time when they'll get their fractional move has
> stayed unpredictable compared with 3.4.3 and earlier, but it
> no longer gets boosted like 3.6.0 had a small chance of doing.)

Not sure what you are aiming at with the 360/36x hint.

The extra monster turns is something I'd have said is behavior that
I thought is - from observation/playing at least - also implemented
in EvilHack. It may make it a bit more difficult but it is (IMO) not
an unbalancing change.[*] This is IMO a sensible change and I don't
mind that at all.

The reason for a choice of one variant or another are several factors
that I already posted about a couple times in the past; I will not
repeat these again now. But in context of the theme of your post, UI
is certainly a very important factor, but not the determining one.

Janis

[*] I posted about some unbalancing parts already. Just now, e.g., I
met on the _2nd mines level_ an antimatter-vortex that disintegrated
quickly my "fabulous" +1 club (no altar, no artifacts, yet) and my
shield and some precious gloves; my only wand of teleportation was
empty, and only by _luck_ (and accumulated ~100 HPs) I escaped that
critter alive.

Keith Simpson

unread,
Mar 29, 2023, 8:46:05 AM3/29/23
to
So this thread has gotten a bit out of control. I came here to RGRN by
request of a few players because they noticed talk about EvilHack
(Janis, to answer your earlier question - yes, I am the dev for EvilHack).
The intent was to field any questions, or resolve any issues/bugs that
may have been discovered - basically an overall 'how can I help'.

I did not join in the discussion here to defend my variant, but human
nature being what it is, it's become increasingly more difficult to try to
not do that. Constructive feedback and criticism is welcome and even
encouraged (and that mindset has contributed to many changes and
improvements in the past), but most of what I'm seeing here is whining
and complaining without any suggestions for improvements, and most
recently, insults. I did not engage in dialogue for this, nor do I have the
time or inclination to continue to do so.

I hope everyone reading can find enjoyment playing EvilHack if you give
it a try. If you're interested in some reasonable back 'n forth, along with
some witty banter and in general a fun time, you can find me in #evilhack,
#hardfought, and several other NetHack related channels on libera.chat IRC.

Loggers VIII

unread,
Mar 29, 2023, 12:06:37 PM3/29/23
to
Janis, I've been watching this unfold from the safety of my beloved #evilhack IRC channel, but I couldn't resist replying to a few of the more ridiculous things. Bad habits, I know.

> That's utter nonsense. It's not about reading it's about a sensible
> user interface! If default values lead to deaths that's bad design.

Utter nonsense is actually a great way to describe this! The thing you don't seem to understand is that you only think it's bad design because you played versions with a different prompt. I started playing with 3.6x and can confidently say that, after playing lots of 3.4.3, the current implementation is better. It makes sense. Press no to stop, press yes to continue. It's simple and intuitive, and even when I had played little of either version I still liked 3.6x's version better, because the older prompt isn't intuitive. If you smash the default input at every prompt, you're not going to do well playing nethack. If you read the message, you will not choke.

The reason this is such a silly thing to call bad design is that 3.4.3 has no bearing on the current quality of game design in NetHack. Why would it? You're not required to play 3.4.3 before playing current versions. The only way this is bad design is if you count 3.4.3 and 3.6x as some sort of homogeneous gameplay experience, which is a very strange way to think about it.

> (Difficulty and balance are completely different things. Slashem,
> for example, is also "hard" but obviously a lot better balanced.)

I'm incredibly curious. Could you give an example? Or three? I have heard a lot about slashem and none of it was praising balance. IMO evilhack is meticulously balanced where it is currently. However, I have not played much slashem, and I'd like to hear someone with your experience's take on it.

And finally, I saw your complaints about how locusts' attacks are impossible to predict. You would be right, but there's been changes to monster lookup. In evilhack you can use / ? to search info of a monster, and it's not just the encyclopedia entry, but it will also show you all the stats and attacks of that monster. I don't blame you for not knowing about this, it's a bit hard to find out on your own.

Overall, why don't you just stop playing evilhack? You don't seem to be enjoying it, you would probably have more fun with something like... well, I don't know. shop around a little, there's a lot of variants.

Janis Papanagnou

unread,
Mar 29, 2023, 5:30:06 PM3/29/23
to
On 29.03.2023 18:06, Loggers VIII wrote:
> Janis, I've been watching this unfold from the safety of my beloved
> #evilhack IRC channel, but I couldn't resist replying to a few of the
> more ridiculous things. Bad habits, I know.
>
>> That's utter nonsense. It's not about reading it's about a sensible
>> user interface! If default values lead to deaths that's bad
>> design.
>
> Utter nonsense is actually a great way to describe this! The thing
> you don't seem to understand is that you only think it's bad design
> because you played versions with a different prompt. I started
> playing with 3.6x and can confidently say that, after playing lots of
> 3.4.3, the current implementation is better. It makes sense. Press no
> to stop, press yes to continue. It's simple and intuitive, and even
> when I had played little of either version I still liked 3.6x's
> version better, because the older prompt isn't intuitive. If you
> smash the default input at every prompt, you're not going to do well
> playing nethack. If you read the message, you will not choke.

No, you are missing the point. It's *not* about the question or
negated question choice, and consequently the necessity of then
either typing Y or N. It's about the _default_ behavior, if you
don't type explicitly Y/N but use the confirmation ('space') key.

I already wrote elsewhere that the first decision must be whether
specific tasks should in the first place allow _defaults_ or not.
_If_ you decide to make default behavior possible it should be a
sensible (and non-fatal) choice!

> [ rest snipped because based of wrong assumptions ]


>> (Difficulty and balance are completely different things. Slashem,
>> for example, is also "hard" but obviously a lot better balanced.)
>
> I'm incredibly curious. Could you give an example? Or three? I have
> heard a lot about slashem and none of it was praising balance. IMO
> evilhack is meticulously balanced where it is currently. However, I
> have not played much slashem, and I'd like to hear someone with your
> experience's take on it.

I try to give an example; Nethack (NH343) had balance issues; you
could early game face a super-fast swarm of soldier ants and just
die, but in the late game monster difficulties and characters did
not match any more. Slashem has increases in monster difficulties
for higher level monster you meet later in the game (dragons and
new golem types, just to give some prominent examples). But it's
also not a black/white issue. Consider for example summonings; in
Nethack they were typically (IMO) already unbalancing, yet more
so in Slashem where exponential growth is just deadly, even with
late-game buffed up characters. For EvilHack (as mentioned) I yet
don't have sufficient experience. For the early game it appears
to me to be yet more unbalanced; example are the locusts swarms
with deadly effects that cannot be countered or healed as often
as would be necessary. Here they added something that I think is
similar unbalanced as early soldier ant swarms in Nethack.

To make a distinction here for clarity; it's not the death attack
or the swarming or the level-depth of their appearance. It's the
combination of all that with the otherwise presented possibilities
to act on such appearances, attacks, ambushes. There's other fatal
new features that I don't count as unbalances; e.g. fungi - was it
black or gray, don't recall - that also can deadly poison players.
But these are more of the "floating eye" sort of threats; you have
contact, probably die, and learn to handle them differently next
time. That's fine.

>
> And finally, I saw your complaints about how locusts' attacks are
> impossible to predict. You would be right, but there's been changes
> to monster lookup. In evilhack you can use / ? to search info of a
> monster, and it's not just the encyclopedia entry, but it will also
> show you all the stats and attacks of that monster. I don't blame you
> for not knowing about this, it's a bit hard to find out on your own.

Wrong assumption. - I know that UI feature (configured it even to
show also their coordinates, IIRC) and use it.

For all the locusts attacks I encountered up to know that feature
didn't help to not die.

>
> Overall, why don't you just stop playing evilhack? You don't seem to
> be enjoying it, you would probably have more fun with something
> like... well, I don't know. shop around a little, there's a lot of
> variants.

First; thanks for your interest in what I am doing (and enjoying).

Of course, and that's what I'm doing. Looking for a (more) perfect
roguelike. I tried EvilHack exactly after I asked to suggestions
and got hinted to it. So if anyone has other suggestions they are
certainly welcome! - The only drawback seems to be that roguelikes
typically seem to be derived from each other. And the convergence
to good design choices seems not a given unfortunately.

Janis

Loggers VIII

unread,
Mar 29, 2023, 10:59:12 PM3/29/23
to
Couple things real quick, then I'm out of here;
About the choking thing. As far as I can tell, you are wrong. Incorrect. Pressing space at the prompt, as far as I can tell, makes you stop eating. If you don't believe me, look at this: https://youtube.com/shorts/2sSYj-kAg04
Second, I had no 'incorrect assumptions', as far as I'm aware. The only assumption I made was that you liked 3.4.3's interface better and were used to it. Everything else is, as I see it, fact.

And again, my 'incorrect assumption' about locusts was based on something YOU said. You complained that there was no way to know that locusts had a sickness attack and that was bad design. You said it was bad because you couldn't anticipate it. And yet, you knew about monster lookup this entire time? that makes those comments blatantly false, and yet you still took the time to complain about an imaginary problem.

On game balance, it seems that your philosophy is that it should only be as challenging as slashem. A tip regarding locusts anyway; they are slower than you. You may have gotten unlucky with the rat king mines level, which has slowing sewage, but then maybe your take away from the locust incidents should be that you oughta get ahold of some ranged weapons, wands, an instrument or a unicorn horn before going too deep into the mines? I think the point of evilhack's encounters is to make solutions to them more difficult (and in my opinion, more fun) to get out of, which encourages creative play using every resource at your disposal. In any case, your criticism once again fails to present an alternative or encourage any intelligent back and forth.

I'm going to stop using this annoying outdated form of communication now (sorry guys it's true), because minor inconvenience might as well be death for me.
Sorry for the messed up voice in the video btw, I'm suffering from an annoying case of strep throat at the moment :(
Anyway, I hope you find the variant for you. Adieu!

Janis Papanagnou

unread,
Mar 29, 2023, 11:47:21 PM3/29/23
to
On 30.03.2023 04:59, Loggers VIII wrote:
> Couple things real quick, then I'm out of here; About the choking
> thing. As far as I can tell, you are wrong. Incorrect. Pressing space
> at the prompt, as far as I can tell, makes you stop eating. [...]

I just tried on a new session on some server; you seem correct.
Not sure what I did to have the game continued in that other
session, since usually I just default my eating to stop, but
in more instances I continued the eating process. It's also
strange that RecRanger posted to have experienced something
similar. - Strange.

>
> And again, my 'incorrect assumption' about locusts was based on
> something YOU said. You complained that there was no way to know that
> locusts had a sickness attack and that was bad design. You said it
> was bad because you couldn't anticipate it. And yet, you knew about
> monster lookup this entire time? that makes those comments blatantly
> false, and yet you still took the time to complain about an imaginary
> problem.

The point is that if I see monsters already known from Real Life
I don't spend time to investigate that. My expectation is that
it's a somewhat reasonable implementation. When you get the first
attack you see the actual concept, but that doesn't mean that you
can do anything about it. I mentioned that my second (or third)
contact with locusts was in the Rats variant of the mines-end
level; at the end of a narrow corridor where they appeared around
the corner and in my back a line of gremlins. Inspecting the
in-game help-screen doesn't help. As said, their swarm appearance
combined with many death-poison attacks they can inflict I consider
not a balanced decision. This is of course another design decision
to the one to give them unprecedented real life attacks. But both
decisions I consider not good inventions, not only for that early
stage of the game.

>
> On game balance, it seems that your philosophy is that it should only
> be as challenging as slashem.

No.

> A tip regarding locusts anyway; they
> are slower than you. You may have gotten unlucky with the rat king
> mines level, which has slowing sewage, but then maybe your take away
> from the locust incidents should be that you oughta get ahold of some
> ranged weapons, wands, an instrument or a unicorn horn before going
> too deep into the mines?

Thanks for the tip. But mind that I have not a hack'n'slash-and-die
playing style. Based on three decades experience, I go forth and back
as necessary. Fight or retreat, doing what appears best to survive.
I don't know about your games, but "your" offensive wands or scaring
music instruments is not what's usually available in my games. In
case of the Rats level I described already above, missile weapons
were not applicable. At some point you just want/need a luck stone
and bite the bullet. It's not so that you have always free paths to
choose from. In one of my recent games the main branch was blocked
by too tough monsters, the first Sokoban level by a fire elemental
(me with 5% resistance), and the mines were not inviting as well.

> I think the point of evilhack's encounters
> is to make solutions to them more difficult (and in my opinion, more
> fun) to get out of, which encourages creative play using every
> resource at your disposal. In any case, your criticism once again
> fails to present an alternative or encourage any intelligent back and
> forth.

??? - Either I don't understand that (mind: non-native speaker) or
it just doesn't match. - If you are asking here how I'd implement
that then it should be obvious that there's more than one option;
one is to remove that death-poison attack, another one could be to
let them appear later, deeper down in the dungeon, preferable in
open areas (as would also be expectable from real life), etc. (Or
combinations thereof.) - The maintainer of that variant should be
able to derive that from any formulated criticism himself and also
decide himself what means would fit best. If he has no clue he can
of course ask. But my experience is that usually they have their
own ideas how any change fits best in their own game-mindset, and
it's anyway not uncommon that criticism is just ignored/dismissed.

Janis

Pat Rankin

unread,
Mar 30, 2023, 5:26:44 PM3/30/23
to
On Tuesday, March 28, 2023 at 12:47:03 PM UTC-7, I wrote:
> [...] The old prompt was "stop eating?" with <space>
> as a synonym for 'no' and there were valid complaints about that
> over the years. The "new" prompt is "continue eating?" and <space>
> is still 'no' but it now picks the safer choice. 'y' and 'n' are reversed

I described this incorrectly. It wasn't that "Stop eating?" was
defaulting to 'n' if you used <space>, it was that it differed from
the majority of other prompts ("are you sure you want to pray?",
"really quit?", "die?"--for explore and wizard mode) which use 'y'
for the risky (or irreversible) choice and 'n' for the safer one.
That's what old complaints were about. Rephrasing it as
"Continue eating?" didn't alter the default behavior (since the
default was swapped along with y and n answers), it was done
to make this prompt more like other confirmation prompts.

To-be-3.7 has OPTIONS=paranoid_confirmation:eat
and evilhack has included that, but it isn't fully implemented.
It works if OPTIONS=paranoid_confirm:eat is used in the run-time
config file, and the 'O' command ('mO' these days) will list it
as a current setting when applicable, but the menu to make
changes when you pick that option doesn't include 'eat' as a
choice for paranoid_confirmation. That needs to be fixed.

Pat Rankin

unread,
Mar 30, 2023, 5:32:30 PM3/30/23
to
On Thursday, March 30, 2023 at 2:26:44 PM UTC-7, I wrote:
> To-be-3.7 has OPTIONS=paranoid_confirmation:eat
> and evilhack has included that, but it isn't fully implemented.

That last bit is not correct; it is fully implemented. This seems
to be a topic that's error prone for me so I'll shut up now....
0 new messages