User:NDusN

Hi from NDusN (unicode variation of теsт), sometimes known as "iddi01", an intermediate-level player (NAO link unavailable until an ascension is done online)

You might have noticed that i play rather slowly. That is because i'm more careful than average (who would have given up their character so easily?) and more importantly an unresolvable BAD connection due to geographical location (it is really starting to annoy me).

I'm quite unique on the fact that i know about the "other side" (Angband and its variants) years before ever noticing nethack, and initially not even vanilla Angband, but ZAngband (probably the Angband equivalent of SLASH'EM, except for difficulty), and i once thought ZAngband was stand-alone instead of a variant!

Misc. stuff

wishless genocideless almost-polypileless non-valkyrie almost-ascension: wishless genocideless almost-polypileless* Orcish Town non-valkyrie almost-ascension**

* 'cause i got nothing useful before my only poly wand explodes for being cursed.

** Seriously "brainlessness" being one of the most common late-game YASDs is not a joke!

my best dig for victory so far: my best dig for victory so far, would have succeeded if only i had not wasted that scroll of taming on those damn mimics! I only take comfort in knowing that other people have done dumber and more regrettable mistakes...

Keep Izchak Alive, Or A Sightseer Will Ruin Your Lucky Day: A NAO Tale

RoundRobin is now a Gnomish Caveman. Or
maybe Cavewoman, I forget.

Two magic lamps in Minetown - wished up GDSM.

Did Sokoban - got the amulet.

You're welcome and try not to screw it up. This character is already an
artifact weapon away from ascendable.

...

I played a little over 2k turns:

Cashed the magic lamp in for the Eye. Obtained a magic whistle and a
tame shapeshifter. Butchered Minetown (a few watchmen survived), then
sacrificed stuff to restore luck.

...

Dude. You killed Izchak! With a rubber chicken! I watched you do it!

There's going to be some kind of insanely bad YASD as a result, just you
watch...

"You hear the rumble of distant thunder..."

...

Something aweful[sic] happened. :-(

Someone apparently unfamiliar with the interface #quit'ed!
Ouch.

...

It was nearly inevitable that a player who is unable to advance a
character to point X on his own, but wishes to experience point X, would
restore the game at point X, and promptly die.

It is rather disheartening that in this case point X was "saving the game
properly".

-- All-time classic from RGRN RoundRobin thread

For "well... see link" fans

Obviously Excalibur can't do nothing by itself, but who was using it? Who?? ;) ;)

Amusingly, Dudley logic applies to Core War as well...

Didn't get the reference?

(not started by me) Discussion on devteam's never-ending nerfing of stuff (transclusion)

< Forum:Watercooler

Background: I have been playing (net)hack since the beginning of the 80's, starting with Hack 1.0.1, of which I still have fond memories of. Frequented rec.games.roguelike.nethack in the 90's, while, contrary to modern times, access to Usenet was common. Was one of the few that ascended all of the classes at that time (without cheating, of course).

Stumbled to

https://nethackwiki.com/wiki/NetHack_3.7.0

which finally prompted me to ask/rant/protest/whatever.

I have been increasingly frustrated about everything getting nefed in this game for over three decades now. Examples:

- Pudding farming. Tried that *once* in the 90's. Got bored really fast and never bothered again. This was a prime example of gain vs. effort. If someone was so dedicated to sacrifice countless of hours of tedious gameplay in achieving some sort of reward, what was wrong with that? Imagine Minecraft nerfing all the farms that exist in that game today? All that would be accomplished by such a nerf would be killing fun for those that use them.

- MC nerf in recent update. OK, I'll just use a cloak of protection then. Which means I'll have to use a gray dragon scale mail, since all the artifacts have been nerfed so that they *all* can be stolen now, and magic resistance is mandatory. Which means I'll have to use an amulet of reflection, since reflection, too, is mandatory, if you don't want all your rings and wands popping up due to some wand of lightning-toting-hobbit. Is this funneling to play the game in the Only One Right Way really fun? I mean, all the other dragon scale mails, amulets, and cloaks are very sub-optimal, right? No room for variation here?

- All the astral plane nerfs that have funneled the player to use the One Right Way To Play of using wands of teleport to clear the path to the right altar. Since the introduction of the planes some 3 decades ago, I have just hated them, over and over again. No fun, just plain tedium of endless messages (that you cannot ingore, lest you miss the all-important "you are slowing down" or "you feel deathly sick" ones etc.). The astral plane is the only one where I die, not because of stupidity, not because of lack of skill, but because that I still try to play the game in some creative way, instead of the One Only Way Of Wands Of Teleport. Fun, eh?

- Nerfs to two-handed weapons. Many years ago they were all rendered unusable because of the curse issue. In fact, there were just a handful that were usable before that, namely the Staff of Aesculapius and the Tsurugi of Muramasa. Both of which were already pretty unusable by their respective classes, because of the risk of them getting stolen... and after the nerf that *all* of the quest artifacts could get stolen, their usefulness dropped way below zero. One of my most memorable ascensions was a rogue wielding the Tsurugi. Was an absolute blast to set up, and fun to play. But today - no such fun any more. As is (or has been for decades), why don't you just remove the 2-h weapons altogether, since they all are utterly unusable by the player?

- Not to mention the unaddressed long-term flaws, for example monster scaling ("which is the optimal level for a character"-discussions in the 90's, a typical red queens race that could be easily solved making the monsters scale by level depth *only*), spellcasting being crippled in all other classes than the wizard (due to the Eye of the Aethiopica being the *only item in the game* that provides mana regeneration, and even wishing for that has been nerfed now) et cetera.

But let's get back to the upcoming version, and I quote from the wiki:

"Covetous monsters may warp to either the upstairs or the downstairs to heal."

- Since this behaviour, meaning teleporting somewhere to heal, even on non-teleport levels, withouht any mana restrictions whatever, is already one of the most infuriating aspects of the game, causing just unnecessary tedium, why on earth should you boost it to heighten the tedium even more? At least, please do something to the infinite mana pool of the monsters, ok? Also, if I understand correctly, if it is not possible to defeat the arch-liches with the One True Method of Sitting On The Upstairs (if they just pop to the other staircase to heal), they are then made virtually invicible. Really?

"Artifacts are made more difficult to obtain by sacrifice; each artifact requires a sacrifice of a certain minimum value."

- Ok, but why, considering most of the artifacts are utter junk, and in fact serve only to add negative modifiers to the game (for example wishing for them, due to the utterly illogical "number of atfifacts created affecting wish chance" nerf).

"The scroll of scare monster no longer works on any @ or unique monster."

- Ok, so you nerf Elbereth, making scrolls of scare monster valuable. But the fact that each use of the scroll costs either a scroll or a potion of holy water isn't enough, you'll nerf this, too. Whatever, but I fail to see the gameplay value (fun factor) in this.

"Alchemy works less efficiently when potions are diluted."

- As if it wasn't nerfed enough already. Read "benefit vs. effort" above.

"The wand of wishing generates with only a single charge, and can only be recharged to (1:1)."

- Considering how the wishing has already been nerfed, I fail to see the logic in here. I mean: wishing for artifacts is a conduct, as is wishing for quest artifacts (or is it? If not, it should be). And how the artifact wish thing and especially the nerfs in quest artifact wishes have been nerfed along the years... sigh.

"Item destruction from elemental sources can be mostly prevented by having the extrinsic resistance to that element or by wearing a dwarvish cloak; failing any external protection, weaker attacks don't destroy as many items as stronger ones."

- Items popping (wands + rings due to electric shock) is already extremely annoying. I don't know what the quoted text exaclty means but if there isn't a 100% guaranteed method of protecting those items, then we'll have a very unfun and unfair game mechanic.

Ok, I could contiunue, but I hope I've made my point. As is, I've tried 3.6.7, and managed to get a monk and a wizard getting killed in the astral plane (trying to find creative ways to manage the tedium there, just to find about everyting being nerfed out from existance, starting from confused create monster scrolls not creating acid blobs to whatever). Hence, if I start a game of Nethack, it will not be a game of the official release, but one of Slash'EM instead.

Sorry, devteam guys, but that's the unfortunate truth. There seem to be lots of good things in the upcoming release, but many of the flaws unfortunately remain. Maybe it is time to rethink if the constant nerfing is worth it?

I haven't been playing long enough to say anything intelligent about the direction of development, but lots of your protests can be mitigated!

- Two-handed weapons: Keep a sack with some holy water and #tip it out if you need. Not to mention, wielding a cursed two-handed weapon is a major trouble, so at worst you can just pray it away. And Cleaver is super fun and viable!

- Can't skip messages on Astral: Set your MSGTYPE to force stop on the messages you care about and escape through the rest.

- Covetous warping in 3.7.0: It's not written clearly, but the way it will work is that every individual monster will pick either up or down and that monster will always warp to the one they chose. For example, in some games Asmodeus will will be up-stairs-warping, and in some he'll be down-stairs-warping. But he'll never warp from the down-stairs to the up-stairs.

- Wishes in 3.7.0: The idea is to make the transition from mid- to late-game less abrupt. Early wands of wishing are being nerfed, but the number of guaranteed wishes is remaining more or less constant (see the orange text at wish#sources).

- Armor variety: 3.7.0 is likely to significantly increase ascension kit variety w/ the new DSM benefits. But even 3.6 is not so punishing as to require absolutely minmaxed kits. For instance MC3 is not necessary, and even MR through artifacts is viable if you have more than one (e.g. Aethiopica + Platinum Card).

- Alchemy in 3.7.0: Diluted potions will have to be dipped two at a time, which does nothing to the expected value but lowers the variance. A little more time-consuming, I grant, but on the other hand the alchemy smock will reduce the chance of a blast to 1/30, which makes up for it IMO.

All in all I agree there's a tedium problem but that's mostly to do with the length of and lack of variety in the late game. Astral I like because it's like a final boss. The wand of teleport is great but so is conflict or taming or jumping or taking it really slow and killing everything. Anyway that's just my two cents :) Gogok (talk) 11:58, 25 November 2025 (UTC)

"Hence, if I start a game of Nethack, it will not be a game of the official release, but one of Slash'EM instead. " More like Super-Lotsa-Removed-Stuff-Slash'EM, since it doesn't have a lot of the cool new features from Slash'EM. --Kahran042 (talk) 20:46, 25 November 2025 (UTC)

Remember that 3.7 is still WIP, and some of the more contentious changes will probably be changed upon player feedback. The wishes shuffle in particular already caused grumbling, since Vlad's new throne is so annoying as to make its wish almost not worth it, and who in the world still needs wishes by Sanctum?

The alchemy nerf will also hopefully be reworked - devteam intent was to compensate for potions of healing being much more common by making the two-step full healing recipe more challenging, but IMO alchemizing full healing is usually not worth it anyway (that's my rant in the link; no idea why Umbrie moved it to 'history' since it's still relevant), so the change will only punish (what remains of) all other complex recipes while still letting players enjoy a lot of max HP from blessed extra healing.

The sacrifice changes evidently try to make the Banes more enticing by enchanting them slightly, and making good artifacts less easily available. It's still unclear how many players will use Trollsbane even if it's +2; again, this may change upon feedback. I feel that in this case, more radical changes need to be ported from some variant to actually pull this off (and boy, there was a lot of such attempts over the years; even I had one).

The item destruction thing is not actually a nerf, you can still use bags or reflection, but fire/cold/shock resistance from equipment also protects your items 99% of the time. It's probably not 100% only to make reflection still have some merit.

As for the 3.6 changes - MC3 was not mandatory back then and it's not mandatory now; all the stuff it blocks is either very niche or can be easily prevented by other means, except perhaps life drain which does merit a change of wardrobe before facing vampires or wraiths, especially if you're luring the latter from the Valley. The Wizard stealing quest artifacts is something that never happened to me, ever, since it's so easy to kill him whenever he appears; I might change my opinion on this when I ascend as a pacifist as pets can't zap death or throw cockatrice eggs, but we'll see.

The only 3.6 change I truly dislike is actually not a nerf - it was putting candles in Vlad's Tower since it defeats the point of having them separate to begin with! I get that with Minetown being bone-able, multi-conduct players might not get them elsewhere, but there are more elegant solutions to this (e.g. revive and restock Izchak on bones load). Tomsod (talk) 15:12, 26 November 2025 (UTC)

Sorry, i'm afraid you're a little too optimistic on the whole "WIP" and "feedback" thing. Based on my past experience, it's the IMHO nature of Github devs to do what they want and largely ignore or decline "feedback", except for obvious bugs. Look at the issues and PRs on Github. Anything balance-related rarely get merged. There is only a tiny chance that a change a dev made will be withdrawn, and you're damn lucky if they even optimize it if the only reason is "feedback". They're still humans, and has their own opinions on balance. And their opinion is The Law on Github and the entire development of nethack.

Look at PR #1443, which makes better sacrifices less likely to yield the weak artifacts to compensate for the limitations on low value sacrifices. 100% reasonable. And intuitive. And got two approvals. And? No dev response after nearly 4 months despite the worthiness. Isn't this proof enough of the stuff discussed in the last 10 or so messages in issue #448.

It is inevitable that every nethack version will get much harder than the previous, especially for conduct players and speedrunners.

Anyway, some important nerfs in 3.7 that the OP missed:

- Stashes. You don't know how important they are until you start using them. If you have a bag of holding from Sokoban then not using them is fine, but else they are like mandatory. The behavior changes of monsters for containers make stashes completely unreliable without paranoid security.

- Natural HP regeneration. A typical XL 14 player with 18 Con has only about 1/3 HP per turn, and even with everything maxed out it's still only 1/2. With reduced food, resting is more costly. Considering the amount of damage players usually take in mid-late to late game, who has that much potions of (extra) healing to spare?

- Helm of opposite alignment. A bunch of commits with descriptions indicating their purpose to be exactly that, totally prohibits ascending cross-aligned with the helm. This mostly matters for speedrunners and low score attempts, but the erinyes part means typos with Stormbringer is no longer that trivial...

- Since recharging wishing add only one charge, uncursed scrolls of charging is now sufficient for this purpose. This is kind of a holy water nerf

- Long sword. Unless you're a knight, you might as well get every fountain dried up before you get Excalibur (i don't understand why samurai are excluded from the knight's exception, they are the Japanese knights, after all, like how ninjas are Japanese rogues)

- Unicorn horn. Yes, lots of variants did the same thing, but non-casters won't come across that much potions of restore ability.

- Sokoban, at least practically. The original muscle memory won't be of much use after the levels got flipped like a coin.

- Completing conducts. Sokoban now an associated "conduct" which is so easy the very existence of it weakens the concept of conducts and makes the relatively easy other new conducts petless and permadeaf look like pacifist. Note it is an achievement in TNNT, which is the right way to handle such things as finishing Sokoban without error.

Of course, there's a lot more, just look around the wiki and search for oranges ;)

I only found nethack starting from 3.6.7, and yet i feel the same as the OP about upcoming changes, and get nostalgic for a version i never played by watching old speedrun ttyrecs attack stuff while standing on repeated Elbereth.

Actually, some of the game's balance issues which the devs or variants attempted to fix using relatively dumb methods, or that has a long discussion on RGRN/whatever since the old times, can be easily fixed by incorporating stuff from the "other side" (Angband and its variants), but apparently nobody thought of that. To name just a few:

- The natural recovery of HP (or mana) appears to be based on the max HP (mana), and Con affects a multiplier on the max HP but nothing else. Beats everything nethack has attempted over the years;

- Monster and object generation is based on dungeon level only. The latter is so great it is tempting to do a variant just to implement that. Think about it: No more GWTWOD or anything like that, no early wands of wishing, yet a much better chance of getting good stuff randomly in late game, etc.

- You can get multiple attacks per round with a single weapon if your Str and Dex are high compared to the weight of the weapon. Doesn't it feel weird that with max Str and Dex and Expert skill you only get one attack per turn per weapon with a dagger, yet a lowly manes get 3 attacks per turn? (ok, non-weapon attacks are "cheating", yet dwarf lords are only difficulty 6 and has two weapon attacks per round with a single weapon).

- Weapons and armor can have misc properties, including a set called "slays". Trollsbane in Angband would be a simple non-artifact "morning star of slay troll". No need to create artifacts solely to create weapons with specialized targets. None of the trouble of boosting the "banes" necessary. (although Angband artifacts could have "slays" as one of many properties they bear)

- If you ever look at ZAngband dungeons, you're so going to laugh hard at the so-called "themed rooms" in 3.7

- "Level feelings" are given to signify the amount of out-of-depth monsters and objects generated at level creation. The one for monsters is given right after entering, while the one for objects is given after a random delay.

- The experience and XL system are, well, sane. You can progress "normally" to max level in a reasonable amount of time.

Again, there's a whole lot of others that nethack is worth learning from. Trying out Angband will give you a whole new angle of view about nethack, much like i, an Angband fan, gets a whole new angle of view on Angband after playing nethack.

Whew, that's a lot of opinions, but i really hope that this post will be put to good use. -- NDusN (talk) 20:07, 1 December 2025 (UTC)

There's a lot of content in this thread, so I probably won't be able to reply to all of it (and some of it refers to events before I joined the devteam). Still, I'll try to get through a lot of it (but without excessive spoilers):

• The reason to nerf strategies like pudding farming is that not every player realises that they aren't required to win. Sure, most people here know that you can win without pudding farming and it isn't significantly harder – but many players come into the game with a mindset of "this game is really hard, so I will need every advantage I can get to win" and then start doing a number of very unfun things to get a minor advantage. Then they end up not having fun, which is an outcome we want to avoid. For tedious strategies, it's really important to somehow clue "this really isn't worth it, you'll have much more fun playing a different way". Removing the strategy from the game entirely is one of the easiest ways to discourage players from using it, and not much is lost in the process.
• One general pattern throughout the first set of complaints is "these changes make it harder to maintain the sort of invulnerability-providing ascension kit that I'm used to". In NetHack 3.4.3, there are some very reliably obtainable and consistent ascension kits that will get you through the entire endgame with no real thought or trouble. This makes the game post-Castle entirely pointless, and a large proportion of players got bored playing it. Any fix to this problem is naturally going to put more pressure on the ascension kit – the ideal is for it to be something which players get to adapt a) to events in the particular game and b) to their own playstyle (the former produces variety and helps prevent the game getting repetitive, the latter allows the player to have more fun, which they can do either by playing a build they want or (if they find it fun) by creating even more variety). So at least from my point of view, the ideal is to pull the ascension kit in so many directions that you can't really have everything, but to adjust the game in such a way that missing pieces matter less and you can work around them. (For example, in recent NetHack, I don't think GDSM + amulet of reflection + cloak of protection is even a particularly good ascension kit – you would find that sacrificing some pieces of that would lead to much easier wins even if things did go wrong on occasion.) I will note that NetHack hasn't quite reached this ideal yet (e.g. playing without MR is currently still a bad idea unless you base your entire build around it – I would prefer it if MRless builds were more survivable and going MRless is something you could reasonably consider doing if you didn't have a good MR source. Of course, doing that would require weakening some of the things that MR currently blocks, and no doubt people would be upset at the further nerfs.)
• The reason to nerf MC from its 3.4.3 levels was that it effectively meant that cancellable attacks didn't matter at all and effectively didn't exist in the game. This post is very consistent in characterising things as "nerfs", but in fact many of them are mostly buffs – in this case, reducing the effectiveness of MC meant that many monster attacks suddenly became a little more relevant than they used to be, and so it was intended primarily as a buff to the monsters rather than a nerf to the players (this is of course a matter of perspective, but there are always two possible perspectives to any change, and a player with the opposite view might well be posting a forum thread "Why do you have to buff everything in this game?"). A very common complaint about NetHack 3.4.3 is that Gehennom is boring there – and one of the reasons it's boring is that most of the monsters can't actually do anything meaningful other than HP damage. We've been gradually improving Gehennom through the past several years, and while it might not be in an actually good state yet, it's a lot better than it was and I think most players would agree with that. Without reducing the effectiveness of MC, that sort of change would be much more difficult to make.
• I don't agree with the comments about the Astral Plane. There are a range of strategies that work in 3.4.3. There are also a range of strategies that work in 3.6 and 3.7. Wands of teleport will generally improve the effectiveness of any Astral strategy, but are far from required and many strategies don't actually need them.
• Two-handed weapons are a lot more usable in 3.6 than in 3.4.3.
• "which is the optimal level for a character"-discussions in the 90's – 30, regardless of your character build. Someone showed me the maths on this recently. (It turns out that a lot of conventional wisdom from the 90s is just plain wrong!)
• "Covetous monsters may warp to either the upstairs or the downstairs to heal." – it turns out that a lot of people using the {{upcoming}} tag don't explain the change properly. There is indeed a change to covetous monsters that could, if you wanted to be misleading, be described like this. The tactical implications might not be what you would expect from the tag, though.
• Potion/alchemy changes: these are addressing a specific exploit. One of the comments above talks about how alchemizing extra healing into full healing for the max HP gain mostly isn't worth it, but seems to implicitly assume that allowing yourself to take another 8 HP in damage is the only use of them. They have another use, though: you can drink them while wounded and gain a lot more than 8 HP. In 3.4.3 and even 3.6, it is very easy to get a stack of 20 full healing prior to the ascension run, and just carry them with you, drinking them when you become wounded. That effectively gives you an extra 8000 HP, which removes almost all the challenge from the game (you don't even need to save them for the actually dangerous levels). In general, drinking full healing for the max HP gain is a waste of a very powerful potion: if people expect that to be their primary use, it's evidence that they're so plentiful that you can afford to waste them on a small HP gain while not even wounded.
• A 99% method of protecting important items is almost as good as a 100% method. Do you really get hit by shock attacks 100 times in a typical game? Are your builds so streamlined that they couldn't afford to work around the loss of one item?
• I also dislike the candles in Vlad's (and for the same reason)!
• Stash security is almost certainly an overblown concern. I've been keeping stashes under the 3.7 rules for a while now, without significant security, and nothing has gone significantly wrong. In fact, across a range of versions (including 3.4.3), I often just stash items by dropping them on the floor unsecured, and although it's gone wrong occasionally, it has never lead to any game-ruining consequences.
• The HP regeneration changes were playtested quite thoroughly. It is intended that you occasionally need to use potions to heal, but as long as you don't play overly recklessly, there will be enough potions. (Another way to think about it is: HP regeneration/food was so plentiful in earlier versions that you could afford to waste the potions purely on boosting max HP, rather than waiting to do that until you're injured – note that potions of full healing still boost max HP in addition to healing your current HP if you drink them while injured (unless your max HP is very high already), you don't have to choose.)
• Helms of opposite alignment – this wasn't my change and I haven't playtested it yet, so I don't currently know what the implications are. If it turns out to be too easy or difficult to ascend cross-aligned, it might be changed again.
• Excalibur – generally it's somewhat balance-breaking for characters to have endgame-quality equipment by the Oracle level. It's hard to balance the threats in the early game so that they're fair both for characters who don't have a top-tier weapon (e.g. neutral Healers) and characters that do (e.g. pre-nerf lawful Valkyries). Knight is one of the weaker roles and has an awkward start, so giving them access to a weapon that can help out doesn't break things as much as giving it to a Valkyrie or Samurai would.
• Restore ability – In playtesting, I usually carry a few of potions of this and usually only need one or two: it turns out that ability drain actually isn't all that common! There are also other ways to restore abilities (not all of which involve restore-ability effects), and being down a couple of ability points for a while usually doesn't have hugely detrimental effects.
• Sokoban – I disagree with the level-flipping. I think players are mostly divided into three groups with respect to their opinions on Sokoban, and it's hard to change because any change that makes one of the groups happy is likely to annoy the other two. Flipping Sokoban levels is one of those rare changes that somehow manages to annoy all three of the groups.
• It's OK to have a range of conduct difficulties, so I don't mind there being some easy conducts. Polyselfless and artiwishless are also very easy.
• I don't read the GitHub page for NetHack, and don't know whether it's official or not. (GitHub doesn't work without JavaScript, which makes doing anything there annoying, and I don't like how Microsoft is trying to centralise all free software development under their control.) In any case, it isn't the actual repository that we use for development. (It is probably good that the repository is mirrored to GitHub as that makes it more accessible for people to see what we're doing, but bad that GitHub gives the impression that it's a good place to submit feedback.) I think some DevTeam members do look at the GitHub issues/pull requests on occasion, but NetHack has historically had a lot of people requesting things that would make the game worse, and it is often hard to sort through them (especially given the limited amount of time that many of us have for development).
• The development version of 3.7 is probably easier than 3.6.7 is; not everyone agrees with me on this but I think more than half of the skilled players I ask do, and less-skilled/new players generally don't know how to take advantage of some of the broken strategies in older versions and thus if skilled players find a newer version easier than an older one, less skilled players are likely to find it substantially easier (although both will of course still be very hard for them). I don't think this wiki currently covers the "3.6 difficulty spike" in much or any detail, but it's a major part of the game's strategy and something that I've been trying to alleviate somewhat for 3.7 (it was, possibly accidentally, created before I joined the DevTeam so I don't know the details, and I didn't find out about it until I started doing substantial playtesting, something which only happened in the last few years, so I wasn't able to fix it earlier).
• 3.4.3 Elbereth spam was kind-of utterly ridiculous and yet many players were using it as a primary method of playing. (There are NetHack bots that demonstrated that pretty much that technique alone could be used to reach Medusa's swamp, even with basically no other knowledge of the game but bump-attacking: if they didn't die to monsters represented by @ or ranged attacks – or team a, which were somehow dangerous enough to even beat the Elbereth spam – they could get down that far sometimes.) As such, it's basically in the same category as pudding farming (players were doing this unfun thing because they thought it would be worth it in terms of ascension rate, and often burning out on the game as a result).
• Tying items to depth is something that we've been considering. It would not, however, be a good strategy to the problem of attack wands as it would mean that the playable character would have no access to attack wands through the early game, when they often really need them. I've recently been trying to collect statistical evidence to see how bad the GWTWOD problem actually is, and suspect that it may still be a significant enough problem to need further changes (there have been some already). So don't be surprised if you see more changes in that direction – but it probably won't be done like you suggest. (Note that Angband uses an entirely different balance principle to NetHack: notably, it uses (or at least traditionally used) non-persistent levels and many of its design decisions were made based on that. For example, level feelings don't make much sense in a game that has persistent levels.)
• Multiple attacks per round are very difficult to balance in NetHack. For example, two-weaponing gives huge penalties and nonetheless is probably the most powerful combat technique in the game.
• The XP/XL curve does indeed go somewhat wrong after somewhere in the level 11 to level 13 range, and I've been considering changes. I suspect the best fix involves increasing the experience yields of monsters rather than changing the thresholds, though.

Ais523 (talk) 21:12, 2 December 2025 (UTC)

I expected the argument that full healing is also good for emergencies, but in my personal experience having good max HP together with good AC mostly eliminates the risk of death from HP loss. I still carry some blessed full healing (naturally-found) just in case, but I hardly ever use it! (and ultimately ascend with several spares) Maybe with 3.7 regeneration changes it will be different, I dunno, but I still feel that the main side-effect of proliferated healing potions you should worry about is how much max HP players will be able to amass, and the alchemy nerf does (almost) nothing about that. That said, I don't see what could, short of completely changing the approach and having the ubiquitous healing items be something else than potions, like SLEX healing scrolls or Brass bandages.

Dying to an early wand of death is something that happened to me twice: one was a bones centaur which still had some charges left, and another an actual gnome. While such a death is sudden and largely unavoidable, my ascension percentage is abysmal already, so I don't feel it's that much of a problem (to me). Tomsod (talk) 04:59, 3 December 2025 (UTC)

Addendum: while it makes sense to forbid farming when it's unfun and unnecessary, I'm rather glad that I'm doing the pacifist run before 3.7, as at the moment I'm about to farm guards for their whistles to polypile into some other tools I want. It may be tedious, but I don't see how I can get these otherwise, and I really need a can of grease right now! Also a wand of speed would be nice, somehow not a single one until the Castle. Tomsod (talk) 05:17, 3 December 2025 (UTC)

On the subject of full healing, you are looking at this from the wrong perspective. Potions of full healing can be used to do two things, let's call them A and B. Your argument is "I've never needed B, and this change doesn't significantly affect A, so it's unnecessary". My argument is "B may be unnecessary, but it's also broken, and therefore needs to be weakened in order to prevent people trivialising the game with it – the size of the effect on A is mostly irrelevant". If you've never used a strategy because it's unnecessary, it's hard to get a feeling for just how weak or strong that strategy is. (Another couple of good examples of similar things: pudding farming and vault guard farming are both entirely unnecessary, and both broken when taken to extremes. If you're just vault guard farming for a can of grease, it might not seem broken. If you're vault guard farming to gain tens of wishes – which is sometimes possible in 3.6 – suddenly it looks a lot more broken.) Another way to put it: if a strategy is both unnecessary and broken, there isn't much downside to nerfing it! I think people have a lot of tendency to view nerfs/buffs based on how they affect their own personal standard strategies, when in fact the change might have been targeted at an entirely different strategy. Ais523 (talk)

Alright, I admit that it's hard not to speak from personal experience when it's almost all the experience I have; I can imagine a mid-game character who can already perform alchemy but may still die to minotaurs or nasties spam, and for them, free full heals (not actually 400 HP because they don't have that much yet, but say 100-ish) indeed remove most of the difficulty there. But there is another part to my argument which I haven't reiterated yet: alchemy has a few multi-step recipes other than healing->extra->full, which aren't that great in 3.6 but might still be situationally useful (mostly for smoky potions). In 3.7 they probably won't be, because this change affects all diluted potions, not just extra healing! Feels like a loss of a feature to me. Tomsod (talk) 15:02, 4 December 2025 (UTC)

Are you thinking of a recipe in particular? Recipes like 1 enlightenment + large stack of booze → large stack of confusion, large stack of confusion + gain energy/level → large stack of enlightenment were never reliable enough to attempt unless you had an extremely large stack of potions to start with, so they're generally only useful in games that have done an excessive amount of farming, and I think almost all the multistep recipes that don't do extra healing → full healing do that instead. The remaining ones go via fruit juice + speed → booze and then dip the booze to get an otherwise useless potion, which becomes not worth it for smoky potions purely from the stack-count changes (so the further effect of the dilution effect doesn't affect it). Ais523 (talk)

Smoky potions are not worth it normally; if you really really need a wish, it becomes more valuable than a couple dozen assorted potions. Like in my current game, I stubbornly spent the entire Castle wand on PYEC (bad move in retrospect, but at least I got it on the wrest), so now I'll be relying on smokies a lot! I imagine such a situation may become more common in 3.7 where Castle only has three wishes. (Also someone might explode their BoH with the wand inside etc. - in general farming can salvage games after a stroke a bad luck or stupidity.) Further, any time you have a stack of diluted potions - either from random alchemy or an unfortunate bath - they might be useful for alchemy in 3.6, but probably not in 3.7. Other than that, turning diluted sickness (again from random alchemy) into booze is a staple of foodless runs, and the change definitely ruins it! (Notably 3.6 did not, as it made an exception for non-magical potions.) Yes, most runs don't care for the assorted alchemy recipes, but they're still part of the game, and 3.7 alchemy change will all but destroy them, without even meaning to. Tomsod (talk) 19:07, 4 December 2025 (UTC)

Update on guard farming: got the can after two batches of 100 kills each. While for a more (neuro)typical player this could've been boring, I found the process moderately pleasant, since every scroll or potion or wand or even well-enchanted armor (for pets) was a welcome surprise! Now, if I wanted to farm enough potions for "tens of wishes" I'd have to spend at least an order of magnitude more effort, and that would be tedious even for me, but the fact that the first hundred killed guards got me every tool imaginable but a can of grease suggests that the 3.7 cap of fifty guards (actually even less, as you'd have to wait thousands of turns for the latter ones to arrive) might be a bit too harsh. Tomsod (talk) 04:34, 13 December 2025 (UTC)