Talk:Speed boots

"20 moves per turn"

"On average, the player will thus get 20 moves per turn (167% normal speed, or 5 moves every 3 turns)." - is either incorrect or extremely confusing - 20 moves per turn means 5 moves every 3 turns? Bulwersator (talk) 18:22, 29 April 2013 (UTC)

(Please put discussion into sections.)

This seems to refer to 20 movement points per turn, which indeed means 5 actions per 3 turns; $\frac{20}{12}=\frac{5}{3}$. Fixed the article. —bcode ^{talk | mail} 18:45, 29 April 2013 (UTC)

Strategy section

The strategy section makes speed boots sound like an early-game crutch item that might end up in your ascension kit. In my experience, speed boots are used by about 90% of end-game characters, and the strategy should reflect that.

I've never seen anyone use potions of speed as a substitute. They're heavy, not particularly common, last only 160 or so turns each, and unbagging and quaffing them takes time. If you do happen to have a decent sized stack of them, you'd probably prefer to alchemize them for more HP. The haste self spell works reasonably if you can get its success rate high enough, but not every character is a spellcaster, and unless you have the Eye you probably have other spells competing for your energy. You'll also probably drop and pick up the Amulet to save energy and nutrition, which, again, takes time and gives monsters a chance to surround you.

So I'd like to rewrite the strategy section to present speed boots as the "default" ascension kit boots, and the spell and potion as alternatives that might work for specific characters. Does anyone disagree? --Darth l33t (talk) 15:00, 25 June 2021 (UTC)

I've been looking at the ascensions happening in Junethack, and I agree that at least 9 in 10 many (for vanilla at least) are using speed boots at the very end. -Actual-nh (talk) 16:30, 25 June 2021 (UTC)

Initial estimate corrected below. -Actual-nh (talk) 16:15, 30 June 2021 (UTC)

I never particularly got that impression - not because I think you're wrong, but because I haven't scrutinized the page enough to really weigh in (that and I have a bunch of other pages I wish to tackle in a similar vein). I'd be perfectly on-board with a rewrite, though. --Umbire the Phantom (talk) 17:11, 25 June 2021 (UTC)

If I remember correctly, I'm the one who made the edits to the speed boots page. I'll make the case against speed boots here because I consider them one of the most overrated items in the game, and are only considered necessary for ascension because the wiki has said so for a very long time. Most of my ascensions do not use speed boots at all.

Contrary to the article, they're a relatively poor wish in the early game. If you have neither magic resistance or reflection, you should wish for one of those properties. If you have one of those, it is optimal to wish for the other to avoid more instadeaths. Thus, you might consider wishing for speed boots only if you already have magic resistance and reflection, so likely the second or third wish in the early game (a rare occasion), where it also competes with half physical damage artifacts.

Later in the game, wishing for speed boots doesn't get any better. A wand of polymorph or a magic marker is far more impactful, since they can be used to get large quantities of genocide and enchant armor to increase your character's HP and AC considerably. Potions of speed are common as defensive items, do not consume a wish, and can be quaffed before dangerous levels like the Astral Plane. Spellcasters can stack thousands of turns of haste self using a power regenerating artifact and never have to recast it for the rest of the game.

I view speed boots like I do gauntlets of power or a helm of telepathy - they are the easily the best armor for the slot, but the advantages that they provide are small. As a beginner, I was often lured into wishing for a full set of these items at the Castle, only to find I was lacking in hit points, valuable rings, and magic markers. Like gauntlets of power, speed boots are a great item to find early on. They are at their most useful at this point since it allows you to gain speed before finding a quantum mechanic corpse or a wand of speed monster, allowing you to easily escape from speed 12 monsters. However, there are much better wishes, and speed boots should not be on the same level of necessity as magic resistance or reflection as the wiki makes it seem. --Luxidream (talk) 09:53, 27 June 2021 (UTC)

I have to wonder if part of this - including some of my observations - is due to the change between versions in treatment of speed making kiting less viable. My observations also can't tell whether people are wishing for speed boots (artifact wishless conduct might also make a difference on this). -Actual-nh (talk) 15:05, 27 June 2021 (UTC)

Thanks for your input, Luxidream. One thing I'd like to point out is that strategy sections should be primarily geared to new and un-ascended players. Against your wishes I checked your NAO profile, and you've played over a thousand games to date, including over a dozen ascensions. Of the three ascension dumplogs I could open, two of them show you missing half an ascension kit, so you're clearly very skilled at playing the end game with sub-optimal equipment. I think you might have forgotten what it's like to be a neophyte overwhelmed by NetHack's incredible freedom of choice and lack of direction, just wishing someone would narrow the scope of options a little. In my opinion, speed boots are by far the best option for new players.

With regards to your rebuttals: "Just cast haste self until you have thousands of turns stacked up." The classic newbie roles are heavy hitters like Barbarians, Samurai, and Valkyries, which have neither the low base penalty to cast haste self reliably, nor the energy to cast it repeatedly. Plus, as you note, they'd need the Eye or the Mitre to have enough energy to pull it off anyway. And unless they luck into a spellbook of haste self, they'll have to wish for one. Rather than spending two wishes on an artifact (which can fail if there's already more than one) and a spellbook, why not just spend one on the boots, which as you note are usually best in slot?

"Magic resistance and reflection are more valuable in the early game." I cover that in my edit, where I suggest (G/S)DSM as a first wish, but it's not unheard of to get e.g. both a fountain wish and a magic lamp from Izchak's. AS for getting both for instadeath protection, that's late-game thinking; in the early game you're more likely to run out of HP in combat than be instakilled. I think speed boots are a fine second wish; DSM may help you endure a leocrotta or swarm of soldier ants, but only extrinsic speed lets you outrun them.

"Potions of speed are defensive items, so kill monsters and take them until you have a stockpile." Yes, and the reason monsters get defensive items is to use them. That potion will be down the monster's gullet before you ever get a chance to get your hands on it.

"Wish for a half physical damage artifact." There is none for lawful characters, and the neutral one weighs a ton (and you can't wish for it as a Valk, and it's a long way to level 14). You'll also take heavy damage when you make the wish, and again if you fatfinger and apply the tool, which can kill you if you're low on HP. And again, if two artifacts already exist, you risk wasting your wish.

"Wish for a wand of polymorph and polypile until you have what you need." Polypiling is complicated, requires a large stock of scrolls and magic armor to get started and to replace what shudders or becomes non-magic armor, and is extremely swingy. You can't count on getting anything useful out of it, so it's a poor option for inexperienced players who need every advantage they can get. Polymorph wands also aren't that rare.

"Wish for a magic marker and nurse dance all night." Again, newbies gravitate toward tanky roles. Nurse dancing gain is based on the ratio of your level to your max HP, so roles with naturally high HP don't benefit as much from it.

Therefore, I still argue that speed boots are a strong early-game wish, behind only DSM. If you want to argue another side, you're free to edit the article, but I think you should keep in mind that newbies will read the strategy section looking for advice, and we want to reduce their confusion, not increase it. --Darth l33t (talk) 16:25, 27 June 2021 (UTC)

I'm going over the ascensions in Junethack more closely now. So far, it looks like that for 3.6+ versions, if someone isn't doing any artifact wishes, they tend to have speed boots. If they did artifact wishes, so far it's roughly 50/50 whether they have speed boots. -Actual-nh (talk) 19:33, 27 June 2021 (UTC)

So Junethack ascensions so far. 3.6+ and based on: ~85% Very Fast (91 out of 107). Others: ~91% Very Fast (41 of 45). -Actual-nh (talk) 23:59, 27 June 2021 (UTC)

For 3.6+ and based on: 12 Artifact Wishless, no speed boots. 14 with Artifact Wishes, no speed boots. 40 Artifact Wishless, with speed boots. 29 with Artifact Wishes and speed boots. 12 with speed boots but unlikely (by Luxidream) or impossible (Wishless or only an Artifact Wish) to be via a wish. -Actual-nh (talk) 01:24, 28 June 2021 (UTC)

I noted that spellcasters could use haste self as a substitute. Neither Valkyrie nor Barbarian are spellcasters. Most players would be able to see that this isn't viable for them from their low Int and Pw. Obviously haste self is not worth a wish either, but if it is obtained without using a wish, such as by writing it as a Wizard or finding it, it can allow the player to ignore speed boots. — Preceding unsigned comment added by Luxidream (talk • contribs)

+ At 173 Junethack ascensions, and 144 usable games, a Chi-square P-value is .03263 for a relationship between Artifact Wishless and having possibly-wished-for speed boots during the EndGame. Proposed change to strategy section: If you don't plan on wishing for any artifacts, then speed boots may be a reasonable wish, although not the first priority. If you do plan on wishing for artifacts, (non-artifact) speed boots are less important than that. -Actual-nh (talk) 15:20, 29 June 2021 (UTC)

Potions of speed are easily obtainable from monsters that can be instantly killed before using them (sleeping gnomes and bugbears) in a throne room or enemies that can't use them (floating eyes, dragons). My realtime speedruns are often able to acquire multiple potions of speed, so it isn't unreasonable for more cautious players to obtain more.

Polypiling often produces speed boots, rings, scrolls, and other items in the ascension kit. Moreover, beginners often have more junk than experienced players do, making this strategy more effective. It is a powerful, reliable strategy when done well. The only runs that don't benefit from polypiling are extremely low-resource speedruns.

While I have more to say on this (see below), it's common for a human Valkyrie or Barbarian to have 100 HP or less at level 14, making nurse dancing still very beneficial to get over 200 HP, a comfortable amount.

Both of the above items are more effective than speed boots. I frequently see newer players killing the Wizard of Yendor with no magic marker, no charging scrolls, and no wishes remaining. A player without speed boots and a magic marker or several scrolls of charging is more likely to ascend than a player with speed boots and lacking anything else. And as I've said, as a new player I often made this mistake, because I was misled by the wiki that speed boots were as necessary as magic resistance or reflection, which they are not.

I don't agree with your philosophy on writing strategy pages. A strategy shouldn't be included because it is the easiest to understand. Should we exclude price identification entirely because it is complicated? I have several wins that do not use price identification whatsoever, so clearly it is not necessary to ascend. In these general strategy articles, we should provide the best advice possible for any character, regardless of class or player skill. If the choice is nuanced, the strategy page should reflect that, and the choice here is not as simple as wishing for speed boots immediately after DSM. --Luxidream (talk) 06:54, 30 June 2021 (UTC)

I removed the section where I compared the alternatives negatively to speed boots; those should arguably be discussed in their own articles or a general strategy article. I still assert that having one item that solves your speed issues for the entire game, with no further expenditure of resources, and competes with little else, is a good wish candidate; and I think the way the strategy section is written doesn't assert that it's a god-tier item, only one that's worth considering. And Actual-nh's data seems to indicate that other players agree with me (I suspect the artifact wishers have the Eye and are haste-stacking). If you disagree, feel free to edit the strategy as you wish. --Darth l33t (talk) 16:02, 30 June 2021 (UTC)

The artifact wishers are a majority Lawful or Chaotic (with the Eye being Neutral); once I get some RL work done, I do want to do an analysis of alignments. I do agree that most (85%+) of those who are not using speed boots have another source of Very Fast speed, with at least half of the remainder having Fast speed - another consideration in this may be whether the character is of a role that gains (admittedly not Very Fast, but Fast) speed; another analysis will be of roles. -Actual-nh (talk) 16:12, 30 June 2021 (UTC)

Thanks. I'll leave the page intact. My opinion on speed boots in well in the minority, so maybe the page should not reflect that. I do wonder how much of that popularity is from the wiki and pre-3.4.3 spoilers more than anything else, though. --Luxidream (talk) 08:41, 2 July 2021 (UTC)

With all 3.6+ (3.6.6, 3.7.0, xNetHack, SpliceHack, EvilHack - GnollHack dumplogs are not available) Junethack ascensions with any wishes, which is a somewhat smaller set (127 instead of 144 total), the p-value is still (using social science standards) significant (.059477). With regard to whether roles gain Fast Speed at some point, this does not have a significant effect with this set of games; I will next check with wizards added to the roles effectively (at least potentially) gaining Fast+ Speed, given the Eye. -Actual-nh (talk) 19:51, 2 July 2021 (UTC)

Not significant checking with wizards considered to have speed available; also not significant looking specifically at whether the eye was available. I did find some rather odd alignment/artifact wishing results - see Talk:Alignment. -Actual-nh (talk) 20:42, 2 July 2021 (UTC)

I think it would be possible to get every vanilla NetHack dumplog from the NetHack scoreboard to avoid having to use variants, which may change the correct strategy. https://nethackscoreboard.org/ascended.nh.html has at least the last 1000 ascensions, and I'd ask the scoreboard maintainer, aoeixsz4, about getting them in a reasonable way.

On your proposed change, I think it should specify half physical damage artifacts. Carried MR artifacts like the Magic Mirror of Merlin and the Platinum Yendorian Express Card are poor wishes that don't compete with speed boots well when it comes to increasing your survivability. --Luxidream (talk) 19:31, 4 July 2021 (UTC)

I can check without any variants for most of the Junethack data; the 3.6+ ones have a total of 17 variant cases (of which 1 is Artifactless - xNetHack); I can try to do that today. With the xlogfiles and some programming, I can construct dumplog links for most servers (em.slashem.me has problems). A bit more programming should be able to extract the speed boots (and fast/very fast) presence. (I do see some cases in which the character ascended - very fast speed - not wearing speed boots, but having them in their possession; I've counted these as with possibly-wished-for speed boots unless there was some reason otherwise. BTW, you named one pair of speed boots "do as I say not as I do"?) I don't know when I'll be able to get around to it, however - lots of stuff is up in RL... sigh. Thanks for taking a look, BTW!

OK re half physical damage (how about half spell damage?). -Actual-nh (talk) 20:22, 4 July 2021 (UTC)

The artifact-wishes vs possible boots wish is slightly more significant without the variants included (0.043389), which considering that it's removing data - which usually would worsen the p-value - is indeed a good argument that variants may have different strategy. I'll check a few of the others. -Actual-nh (talk) 02:54, 5 July 2021 (UTC)

Sometimes with extra wishes on a speedrun, I will wish for speed boots on the ascension run to reduce harassment and save time, especially if I've accidentally lost intrinsic speed to a shade. I'm poking fun at the fact that I normally advise against wishing for speed boots, which I still stand by. Many speedrun wishes, such as "3 cursed genocide", "amulet of reflection", and "3 cursed gain level" would be bad in a normal run, and I use speed boots for a similar purpose there. --Luxidream (talk) 05:56, 5 July 2021 (UTC)

At 173 Junethack ascensions, and 144 usable games, a Chi-square P-value is .03263 for a relationship between Artifact Wishless and having possibly-wished-for speed boots during the EndGame. Proposed change to strategy section: If you don't plan on wishing for any artifacts, then speed boots may be a reasonable wish, although not the first priority. If you do plan on wishing for artifacts, (non-artifact) speed boots are less important than that. -Actual-nh (talk) 15:20, 29 June 2021 (UTC)

I will redo the analysis with a few more ascensions, and also try with or without having the pre-3.[67]-based games not included. -Actual-nh (talk) 05:51, 30 June 2021 (UTC)

Admittedly, it is true that "correlation != causation". I'm having problems figuring out another explanation, though. Both of them (artifact wishes and no speed boots) being from a common source, namely playing style, leads to the same advice. -Actual-nh (talk) 05:58, 30 June 2021 (UTC)

Sorry, I was in the middle of editing the page, and didn't mean to publish. Can I retract that comment? Thanks. --Luxidream (talk) 06:48, 30 June 2021 (UTC)

To clarify, I was working on a lengthier comment to address the points above. I'm inclined to believe your analysis. --Luxidream (talk) 06:50, 30 June 2021 (UTC)

Understand! -Actual-nh (talk) 15:27, 30 June 2021 (UTC)