<- 1  -   of 26
List results:
Search options:
Use \ before commas in usernames
I've been looking through HotU, at least partly from a speedrun point of view. (Also I tried to work out how the ending works. My main conclusion is that it doesn't :-P)

One thing I noted is that AFAICT, the only familiar clip is in Chapter 1, right (and it's a nonstandard sort of clip that exploits an oversight in the scripting, rather than a "true" clip)? This implies that we might be able to route out the Wizard levels entirely. So I looked for the second-fastest way to solve HotU Chapter 1 Level 1. I'm pretty sure that solving the control rod puzzle is faster than the alchemist's fire route (because that route requires you to run to both ends of the map); unlike what some casual players might think, the puzzle can be solved with just two rods (most but not all combinations work), and there are multiple places you can find each rod. In other words, the level works a bit like chapter 2 of the original campaign, with a variety of possible solutions and any% being a very low completion percentage. You can get two rods by randomly teleporting around if you're (very) lucky (whenever you take a teleporter, it leads to a random different teleporter, rerolled each time), but unfortunately it's not a combination of two rods that can solve the puzzle. However, there's a room fairly near the entrance which contains a white rod and a yellow rod in adjacent chests (possibly this is a mistake by the developers?), and that is a combination that can solve the puzzle. To reach it, turn left at the pillars, follow the corridor round to the right, and then use the door on the left that goes to a different area. The second room of that area contains the chests. (I've actually made a list of all the rod locations, which I can post if anyone's interested; I'm pretty sure white+yellow is the fastest combination, though.)

Unfortunately, it's not quite trivial to pick up the rods with an "unusual" speed character; one of the chests is unlocked, but the other is locked and has hardness 40 (!). However, it also has only 20 hp (meaning that elemental damage can destroy it without too much effort, although slower than I'd like), and its lock is vulnerable to Knock or to Open Lock (DC 28; as there are Thieves Tools +10 in the Armory, and 20 (base value for skill use outside combat) + 10 (from the tools) + 1 (minimum Open Lock skill) - 2 (minimum Dexterity) is 29, any character with lock opening ability can pick the lock if they can avoid being in combat at the time (the room is, unfortunately, full of enemies). Incidentally, if the character has a lot of Open Lock skill (i.e. maxed), they can plausibly pick the lock even inside combat. Something I haven't checked is whether Sharwyn can be persuaded to pick the lock, but getting her to do that while the room's full of enemies seems fairly implausible.

Part of the reason I've been testing this is that, unlike in the original campaign, it seems reasonable that a lot of time could be saved at the start of HotU via the use of a less custom character; level 1 is free, but levels 2 to 15 are running on the timer. Being able to use autolevel would be nice, and being able to use a pregenerated level 15 even nicer, as long as there isn't too much of a cost incurred by the suboptimal character choice later on. The most obvious choice is Rebekah Stormcoast, who's a level 15 monk (and thus runs a little faster than a level 14 or 12 monk with a familiar would). I've confirmed that she's at least capable of clearing Chapter 1 by herself in what looks like a plausible speedroute (making a detour to pick up and use the rods, but otherwise using basically the same route as in the HotU WR Mejcel linked earlier in the thread); and she does it by brute force (opening the chest with sonic damage from gloves, and just tanking hits rather than being invisible). However, I didn't get anything like a good time (door clipping while enemies are attacking you are a pain).

More plausibly, it's probably possible to do something like this: build a monk 1; start a game of HotU; buy a Wizard level; export the character; restart a new game of HotU with the exported character, and choose to autolevel to 15. I'm not 100% sure, but I think that would buy 14 monk levels automatically, and configure based on package. There might be a monk package that has a good set of skills.

(Silly idea: start a game of HotU; level up the character, automatically or manually; export the character; then enter it in one of the other two campaigns, and cruise around with a character that's 14 levels overlevelled. I'm suddenly really curious as to whether this would comply with SDA rules.)

Meanwhile, I doubt Chapter 2 is significantly improvable (unless it turns out that there's a faster way to get a cutscene including the Valsharess). Sudden realisation: is it possible to kill the Chapter 1 Valsharess, and what happens if you do? (My guesses are "no" and "nothing", but I'm sufficiently non-confident in them that this has got to be worth testing; there's at least one vaguely plausible way to screw it up that would skip directly to chapter 3, which would at least make the experience route very interesting.) Interestingly, the glitched Chapter 2 Valsharess boss fight is one that never comes up in the intended gameplay (and in fact, it's significantly more difficult), so speedrunning has managed to create a rather fun and interesting challenge accidentally. (AFAICT, it's in the same room that the intended fight is in, involving the same Valsharess, but the room is changed from its cutscene layout to a 1-on-1 battle layout during the intended sequence of the game, whereas the glitched fight is inside the scene from the cutscene.) The other thing that I learned doing a HotU testrun is that the cutscene warp glitch is much more difficult than the video makes it look; I can do it, thus confirming to anyone who might potentially be sceptical that the glitch is a real one, but it took me way more tries than it should have done.

Chapter 3 does look somewhat improvable, though; there's a lot of running around in the HotU record video that doesn't seem like it should be strictly necessary. If it weren't for the gold grind, I'm fairly sure the fastest route would be to go straight to the temple, and then just outright kill the Sensei and take her Amulet. (With the right dialogue path, the Amulet's enough to unlock the path to the Puzzle Ring if you have sufficiently high Wisdom. Getting Wisdom high enough should be trivial given our access to epic-level equipment, if it doesn't make it manually; IIRC, the required value is fairly low.) There should be only a minimal need to mess around in the Cania hub area.

As for the gold grind itself, I'm not sure how necessary it is, but I'm also not sure if killing the final boss could be done at all quickly (apparently, the developers were aware of the normal speed kill trick, as many of the final boss's immunities seem to be targeted at the quick kills specifically; why else would it have effects as narrowly specific as Immunity: Movement speed decrease and Spell Immunity: Harm?). If we turn out to need the 700k gold to buy a sequence break, then I think the fastest way to do the later stages of the grind will be to buy as many of the item we're using as we can afford, then sell them all at once. (When testing out the gold grind, I used a highly enchanted plate armour which the blacksmith in the inn (who was also in Chapter 2, but has a higher gold limit now) sells in infinite quantities, and which produces quite a lot of gold per round; that means you don't have to go free Lazy at all. I don't know whether or not that's optimal.) I think I'd rather try to find a "quick kill" rather than needing to grind 700k gold, though; given the levels we're grinding to anyway, a kill is reasonable enough, but whether we can kill faster than the gold grind is another matter. Bonus points if it can be done only with Monk levels, or with Monk + a 1-level splash.

Incidentally, I also did a quick Diamond run of the OC in order to unlock everything it unlocks, and attempted to use my unpatched strategy. It didn't really go too well (because too much has changed), and in particular, I ended up with a completely nonsensical and unfocused character. The interesting thing is, though: I still won, despite the character making no sense (it was something like level 15 or 16 with the levels split between Wizard, Rogue and Barbarian in a completely suboptimal way). This makes me wonder if I'm currently overgrinding on unpatched OC (although one issue there is that even though the extra grind doesn't seem to be necessary to make things possible, it may nonetheless make them faster). It also makes me think about the idea of a minimum-experience run again; at this point, I suspect the biggest sticking point will be Maugrim (as most of the harder fights are skippable, but the requirement to kill Maugrim seems to be pretty ironclad; note that with all the ways to cheese the Morag fight, it can't really be considered difficult at this point).
Edit history:
Mejcel: 2017-01-03 09:47:00 pm
Mejcel: 2017-01-03 09:46:16 pm
Sounds like you're almost planning a run other than any% for Diamond OC? Smiley

About the familiar clipping mechanics, every spot seems to be unique. Thickness matters a lot, as well as how far you can get "inside" a material you're about to clip through (e.g. ~all doors). Also, if there is too much space near your character, the familiar is going to push you to that space instead of clipping you through, so you usually need to be cornered somehow. Interestingly, the Aqueducts clip teleports you all the way to the bottom of the bridge, but I haven't really seen other clips teleport you so far in Diamond.

You are able to clip vertically as well, see for example this video, but it's rare and for example the ledge at the start of Peninsula is something I have never been able to clip through.

There's also the trick of summoning a Familiar to hard-to-reach places, but I haven't found a use for the glitch (other than skipping Undermountain level 1 as you saw), as there seems to be no way to teleport your character to a Familiar or enter a new area with a Familiar.

The question is: are there still possible familiar clips in OC that Diamond runners have never explored?



I won't go into exporting characters or using pre-generated characters etc. myself, however:

1) Your (even partial) automatic level-up at the beginning sounds interesting. Leveling up takes me 38 seconds at the moment. I could start as level 1 Monk, level up Wizard, then autolevelup the rest. This led to me thinking:

- Is level 3 Wizard even necessary? It's only used for Invisibility. Maybe Invisibility isn't that useful anymore in the run? This would also mean I never need level two spells -> 12 Intelligence isn't needed, although 11 is required for me to be able to cast level 1 spells (I think True Strike is necessary @ Valsharess).
- The cons: I would lose -2 Strength, Blind Fight and Great Cleave (although I could of course take the feats later on). I wouldn't have a safe amount of Appraise for the first merchant and I couldn't identify the dagger and the Blue Ioun Stone at the beginning. I would lose all my current skill points at level 15 to useless skills such as Move Silently etc. But maybe do it only partially, and hand-pick some skills/abilities/feats? I don't know.

2) Killing Chapter 1 Valsharess (or rather, Matron Mother) does nothing. About the Chapter 2 warp glitch: it's easy when you figure out the timing (which is at least for me just when the loading screen is over).

3) The thing with Chapter 3 is, the Mephistopheles fight takes like 5 minutes longer than doing all the work to learn his True Name. It is a multi-part fight with nothing you can do to speed it up as far as I know, and even then you're going to need a good amount of money for at least the equipment to fight him.

4) I use the Eyes of Petrification because its sell price with perfect Appraise (49976) is the closest to the maximum buy price of the Chapter 3 merchants (50000), thus also netting you the most money per one cycle of sell-buy (6247). However, you're probably right that buying items with infinite amounts could be faster. This leads to several points:

- Is it faster to buy with all the money you have, or only enough to fill one inventory page? Using for example the Full Plate +6 (48017) you get only 10 items per (already empty) inventory page.
- You still have to go to the Quarry Office and do the quest to open the demon's store. At the very least you are going to need to buy Potions of Heal from him, but the +10 STR belt and the gloves +10 are also basically necessary fighting Gilgamesh and the last Guardian of the Path.

5) Still, finding a way to skip Mimic would be great, but I'm not aware of any. There's just too many places the Demon Hand is used for in the latter parts of the run, and finding a way to Familiar clip all of them.. unlikely.
Edit history:
Mejcel: 2017-01-05 07:47:28 pm
Mejcel: 2017-01-05 07:47:21 pm
Mejcel: 2017-01-03 10:14:33 pm
Mejcel: 2017-01-03 10:14:14 pm
Mejcel: 2017-01-03 09:56:15 pm
Mejcel: 2017-01-03 09:44:28 pm
Mejcel: 2017-01-03 09:43:30 pm
Mejcel: 2017-01-03 09:38:19 pm
Mejcel: 2017-01-03 09:36:28 pm
Mejcel: 2017-01-03 09:35:31 pm
Mejcel: 2017-01-03 09:34:41 pm
Mejcel: 2017-01-03 09:32:39 pm
Just to clarify the deal with Mephistopheles - I did some test runs. Let's say we were to fight him:

1) We save 700 000 GP for not having to learn Mephistopheles' true name.
2) However, we're still roughly 75 000 GP short of getting all the equipment we need if we don't do the Appraise trick at all (+10 STR belt, +10 ATK gauntlets, around 8 heal potions PLUS 4 extra for Mephistopheles). But - it's possible to downgrade, and buy +10 STR, +7 ATK, 12 heal potions with the money. Thus there would never ever be a reason to go to Hellbreath Tavern and we wouldn't need to perform the Appraise glitch.
3) This would take around 1m17s (from the City of Lost Souls to the Temple of the Sleeping Man), saving 2m37s in City of Lost Souls compared to my WR run.
4) It takes roughly 3m30s to fight Meph (and 4 extra Potions of Heal).

It's still a ~minute slower to fight Mephistopheles.

5) However, maybe it's worth it to downgrade the +10 ATK gloves for +7 ATK gloves anyway, and generate less money? The last Guardian of the Path has 39 AC, my attack bonuses are +47/+44/+41/+38/+35, minus 2 to all attacks from Flurry of Blows, minus 3 from worse gloves. So +42/+39/+36/+33/+30... I don't know. But that's the only fight I benefit from having the better gloves, for comparison, Meph has an AC of 28 and Gargamesh an AC of 31.

EDIT: tested #5, it would save 36 seconds. It's totally worth it.
I guess my interest in HotU has mostly been in routing rather than running.

Doing the early stages without invisibility is actually really frustrating, especially Undermountain Level 3 (which requires either taking a detour to trigger Nathyrra like you're supposed to, or else multiple door clips while enemies are trying to kill you). It's definitely possible, but there are way too many enemies who do inconvenient things. It'd totally be worth it to rely on potions if you could buy them conveniently, but I don't think you can; spending three levels just for one spell might actually be the fastest option.

I actually find the familiar "clip" on Undermountain Level 1 really confusing. I haven't managed to clip the familiar directly to the goal yet, although I have managed to clip myself far enough into the pillars that I can then quickrest to regain the familiar and clip the familiar to the other side. This could just be because I'm unfamiliar with Diamond clipping mechanics; it's so uncomfortably slippery compared to unpatched. (Incidentally, with respect to the Aqueducts clip in the original campaign, it feels like there's a mistake with its coordinates somewhere; it's like the game is doing something like drawing areas far apart on the map but considering them to be close internally. I've had instances where saving on the bridge and reloading has placed me behind the entrance. Would be great if you could somehow just save and reload to end up on the other side of the gate, but I've never had that happen.)

In terms of Appraise glitching, I've been filling a page and then selling the page again; that seems like the fastest pattern. The sale limits are really annoying because they pretty much defeat any chance you'd have of quickly gaining money by other means (it's not like you can go out of your way for a single valuable item, like the segmented run that started of this thread did).

HotU Chapter 3 seems distressingly nonbreakable simply because it's so linear (much like chapter 4 of the OC, come to think of it). My main interest would be in somehow finding a "quick kill" on the final boss, rather than fighting the intended way, but none of my usual tricks work; it's like the developers went and found someone who was familiar with speed tricks in the game and asked them to make sure that their final boss wasn't cheesable. One of the possibilities I haven't checked is death magic; I'll need to check what immunity flags there are against that, and whether any of them might potentially be bypassed by the right spell. (Unfortunately, we're probably talking about needing a natural 20 or even a double natural 20 to hit with it, so it wouldn't be faster except in a segmented run.) As far as I can make out from the scripts, though, you could skip right to the end of the fight by managing a kill that wasn't based on HP damage.

Incidentally, I managed to use one of my multiplayer mode tricks (probably not usable in a run due to being so slow and questionably rules-legal, although they can duplicate non-plot items such as piles of gold) to sneak the Relic of the Reaper into Chapter 3. There's special code for that, implying that the developers were originally planning to have it available, but changed their mind. It sort-of works, but not very well: you can place a portal, and return to Gatehouse via the relic, but apparently not return to a portal from Gatehouse (unless Gatehouse was just screwing up for unrelated reasons; I sometimes have trouble using the doors even in chapter 1 and 2); and although the code for handling post-respawn exists, the game intentionally softlocks when you die (forcing you to reload a save) and thus you never get the opportunity to test it. Additionally, setting foot in Gatehouse, while holding the Relic, removes the Relic from your inventory; and I can't really see a way to save time with it even if it had full functionality. So I'm just mentioning this as a discussion point in case it gives you or someone else some ideas.

With respect to the +10 STR belt, I have a mild suspicion that there's a guaranteed one as a drop from an enemy somewhere in Chapter 3, but can't remember where. (Of course, it's possible it's just a random drop, but much of the Chapter 3 loot is predetermined.) Maybe I'll have to go look for that.

Re: any%, that was more a matter of rules questions and "hang on, what restriction is there on characters anyway?" The rules questions are being discussed here.
Might be magic...
I love watching all the new tricks appearing in this thread. If I knew the game well enough, I'd try to find some myself. Have been trying to think of something useful, but came up empty. Good luck guys!
Edit history:
Mejcel: 2017-01-07 06:10:10 pm
Mejcel: 2017-01-05 09:29:55 pm
Mejcel: 2017-01-05 09:12:40 pm
Mejcel: 2017-01-05 09:11:30 pm
Mejcel: 2017-01-05 09:10:42 pm
Mejcel: 2017-01-05 09:08:56 pm
Mejcel: 2017-01-05 09:06:48 pm
Mejcel: 2017-01-05 08:40:07 pm
Mejcel: 2017-01-05 08:16:07 pm
Mejcel: 2017-01-05 08:10:49 pm
Mejcel: 2017-01-05 08:10:34 pm
Mejcel: 2017-01-05 08:10:14 pm
Mejcel: 2017-01-05 08:09:36 pm
Mejcel: 2017-01-05 07:53:50 pm
Mejcel: 2017-01-05 07:52:34 pm
Mejcel: 2017-01-05 07:52:12 pm
Thanks gammadragon Smiley

About Invisibility, it's absolutely needed at least in the Mimic's Nest and Puzzlers' Sepulcher. However, there's an option to buy Potions of Invisibility from High Wizard Gulhrys in Chapter 2, preferably after placing the Valsharess warp, as then you lose no time backtracking: just use the Relic of the Reaper and continue as usual to the throne room.

Having looked at the scripts, I'm doubtful you can find a way for a quick Mephistopheles kill, as he seems to have an Immortal flag on at all times.

I found an interesting strategy for the automatic level up, but it presents some new problems. It involves starting as level 1 Monk, manually leveling up to 9 (8 Monk, 1 Wizard), picking the feats Blind Fight, Weapon Focus (unarmed strike) and Great Cleave, and on the 9th level taking 6 Appraise and 12 Lore. Then I hit the auto levelup.

- 10 second faster leveling up in the beginning
- Gain all the feats I originally wanted (computer picks Toughness, Improved Critical (unarmed strike)

- Lose 1 Strength = Lose 1 ATK, 1 damage (computer picks Wisdom, I don't think this is a big deal)
- Unable to identify the Lesser Ioun Stone, needs a Lore of 16 (not a big deal, you lose 100 GP and perhaps <1 second identifying in store)
- No Invisibility
- Can't use Expeditious Retreat at the very beginning of the game (lose 1-2 seconds I think)
- Real possibility of failing to have enough money for the first merchant White Thesta

This actually leads to two very interesting choices, as having only 1 Wizard level also opens up the possibility to get to level 35. I'm really torn between the two, here are their pros and cons:

- Manually level up to 11 (8 Monk, 3 Wizard), then hit auto
- Start the game as (12 Monk, 3 Wizard)
- At Halaster, level up to 34 (30 Monk, 3 Wizard, 1 Rogue)

- Save around 6 seconds leveling up
- Still got Invisibility
- Overall less RNG in the beginning (compared to option 2)


- -10% Monk Speed (compared to option 2)
- -1 Strength (I don't think -1 attack & -1 damage is that much of a con though, really)

Time saved (compared to WR run): 6 seconds

- Manually level up to 9 (8 Monk, 1 Wizard), then hit auto
- Start the game as (14 Monk, 1 Wizard)
- At Halaster, take 10 extra seconds to XP glitch to level 35 (33 Monk, 1 Wizard, 1 Rogue)

- Save around 10 seconds leveling up
- 10% extra speed from level 33 Monk(!)

- -1 Strength
- Lose 10 seconds at Halaster
- Lose 6 seconds buying two Invisibility potions (buy them from High Wizard Gulhrys after placing the Valsharess warp)
- Lose 2 seconds leveling another level (no feats or attributes to place)
- Lose some time having to identify Lesser Ioun Stone / not having Expeditious Retreat in the beginning (2s? 3s?)
- Have to "yolo" Undermountain without Invisibility
- Overall more RNG in the beginning

Time saved (compared to WR run): ~ -10 to -11 seconds plus whatever +10% Monk Speed gives you

Is option 2) better? Is there any way to measure if the extra 10% speed would be worth the supposed difference of 16 to 17 seconds? Undecided

EDIT: Found out that no matter if it's option 1) or 2), I have an unnecessary rest in Undermountain in the WR run. Previously thought option 2) had it better.
I've made "racetrack" modules before now to test running speed (they're nothing but a long empty room, plus triggers that time how long it takes to run from one end to the other). That was in unpatched, but it should be easy enough to recreate for Diamond. The problem is, although it's possible to measure how much faster a level 33 Monk runs than a level 30 Monk, it's much harder to measure the amount of running that's necessary in Chapters 2 and 3.

Option 2 seems nasty in terms of getting through Chapter 1; the yolo is definitely possible, but I suspect it might lose a substantial amount of time due to having to react to enemies and the like (especially because the have a habit of casting Darkness when I'm trying to do a door clip). That said, it might also be a lot of fun when it comes to movement technique. I can see why you use the mouse; keyboard movement doesn't work when you're going at Diamond Monk speed, because your turning circle is too large, whereas the mouse can turn corners instantly. My problem is that I'm much less accurate moving with the mouse than I am with the keyboard (as my playstyle is generally to menu with the mouse whilst moving with the keyboard, and in unpatched there's rarely a reason to change that), so when I play Diamond I'm struggling with even the most basic movement techniques.

My guess is that Option 2 is probably faster (although I'm unsure), but either way it's going to be less fun to run and therefore more fun to watch. I agree that it's likely to be very tight on cash; something I keep forgetting to do is to sell the armour I'm currently wearing in order to afford the Robes of the Dark Moon, and even then, it's not nearly enough by itself (you need to loot almost everything of value from the Armory).

I think I'm likely to spend most of my HotU-focused effort on glitch-hunting. Windswept Battlefield seems like it might be potentially breakable, for example. (Actually, another thing that crosses my mind is to try to skip the item loss at the start of Chapter 1 somehow. This would be potentially relevant if you were using a level 15 pregen, as they come with fairly good items. Obviously, on a level 1 pregen, it's pointless.)

Incidentally, on the subject of fun/joke categories, I came up with the silly idea of taking one of HotU's pregenerated characters (which are level 15, and available even from a completely fresh install) through the OC. The monk, Rebekah Stormcoast, lets you have monkspeed right from the start of the game. Perhaps more surprisingly, she comes with Boots of Speed (of all things), thus giving you haste right from the start of the game too. This pretty much has to count as a separate category, and probably isn't all that interesting, but it is amusing (and not as trivial as it seems, because many enemy groups in chapter 1 scale to your level; most areas are still relatively easy but Peninsula is an absolute nightmare).
There's no need to worry so much about Chapter 1, it's still doable. I guess it's down to actually testing it out exactly as you said with a custom module, and then just compare it with a video of a run trying to estimate if it's worth it or not.

I didn't even know about those pre-generated characters since I deleted them on day one from my install of the game Smiley
Edit history:
Mejcel: 2017-01-07 08:14:00 pm
Mejcel: 2017-01-07 08:11:50 pm
Mejcel: 2017-01-07 08:11:25 pm
Mejcel: 2017-01-07 08:10:54 pm
Mejcel: 2017-01-07 06:25:25 pm
Mejcel: 2017-01-07 06:24:44 pm
Mejcel: 2017-01-07 06:22:43 pm
Mejcel: 2017-01-07 06:22:35 pm
Re-timed buying the invisibility potions, it's 6 seconds instead of 8.

I actually timed the running times between level 33 & level 30 Monks using just segmented savefiles. It's not perfect of course - there might be some variance and errors - but we get a rough estimate. End result: you save ~11 seconds by having 10% extra speed.

I also triple-checked things: practised a bit leveling up in the beginning and got these results:

- 32,71 s fastest level-up with 8 Monk/3 Wiz (Option 1)
- 25,90 s fastest level-up with 8 Monk/1 Wiz (Option 2)

Difference: 6,81 seconds.


With option 2) you lose 10+6+2+3 = 21 seconds (Halaster, buying Invis potions, one extra level up, have to identify Lesser Ioun Stone, no Expeditious Retreat), but gain around 11 seconds from the extra 10% speed. Thus you lose only 21 - 10.87 = 10,13 seconds.

10,13 > 6,81, thus option 1) is better by a few seconds.

If something seems wrong, please correct me Smiley
I tested a few more things.

The first thing to note is that I've now completed SoU casually, and have been having some ideas about that. In particular, a lot less of Chapter 1's plot needs to be done than it might seem. I've verified (using Rebekah Stormcoast) that the minimum you have to do to complete the module is to go directly to and kill Tymofarrar, do not pick up the mask, then go to Blumberg and give Deekin his skull (and looking into the code implies that even if you don't pick up the skull, Deekin will realise that Tymofarrar is dead and believe a claim to that extent). If you pick up the mask, Deekin will despawn until the J'nah quest is complete, which is clearly a massive detour. The reason that this works is that the only artifact you actually need is the tower statue (in fact, it's intended that you can sell the others!); the game compensates by arranging things so that a casual player will almost certainly get the tower statue last (and thus what happens to the others doesn't matter), but it's possible to not get the others at all. (In this way, it's fairly comparable to chapter 2 of the OC, in which the obvious casual route and the speedroute have very little in common, because the casual route is designed to lead you down the least fruitful paths in terms of completing the chapter first.)

Of course, the big problem here is reaching and then beating Tymofarrar early in the game; reaching him is hard enough as it is (although you might be able to do something stupid involving invisibility or Hide, you'd also need to deal with traps, ideally in a faster way than bringing Dorna along), and beating him without the Ice Phylactery seems incredibly difficult. My suspicion is that the fastest route involves doing quests (either sidequests or main quests) in order to grind up experience (I think you get Phantasmal Killer at Wizard 7; that might be the correct option if there's a fast way to grind experience, but maybe there's a faster one); also that neither getting the Ice Phylactery nor killing J'nah is part of the fastest route (because that dungeon is really long and not really amenable to the normal methods of skipping things). The fastest way to get the first couple of levels is likely via solving sidequests through dialog (meaning that buying Persuade points may be the best option in terms of aiding grinding); some of them give pretty large experience payouts.

In HotU news, I was considering the possibility of skipping the boss just before the Knower of Names, via clipping around the trigger that normally starts the fight. The good news is that the trigger is fairly easy to skip via grappling over it (there's a conveniently placed grapple point). The bad news is that the game considers the fight to still be ongoing if you do that, and refuses to let you interact with the catapult as a result; so unfortunately I don't think it's possible to save time like this. (Something else that I didn't know, but you probably did, is that it's possible to skip the Grimgnaw/Maugrim/etc. fight via running diagonally backwards and then working your way around the wall for a bit; this entirely avoids stepping on the trigger that starts it. This is similar in effect to DCR'ing the fight, but even stupider, and almost certainly faster than trying to clip out of the room with the fight active.)
Edit history:
Mejcel: 2017-01-11 06:19:33 pm
Mejcel: 2017-01-11 06:18:28 pm
Mejcel: 2017-01-11 06:16:38 pm
Yeah, it was known to me that all you need is the Tower Statue, I think I mention it in my commentary as well, but there's no death magic available early on and there is to my knowledge no XP glitches available in the game either. I ultimately ended up killing J'Nah and leaving Tymofarrar alone.

It's faster to kill townspeople than do a lot of the quests since the XP rewards from the quests are usually not that great. I'm not sure if you should kill more people than I did in my WR run though, maybe it could still be faster than the few quests I ultimately ended up doing. Here's a link to it if you aren't aware of it yet, maybe you can find some improvements.

Yes, skipping the Grimgnaw/Maugrim fight was known and has been there in the runs for quite a long time.

By the way, I got a Youtube comment which is quoted below:

Quote from Noggaholic:
I've been playing NWN on and off since release and have only just discovered that it was being speed run, something I never would have thought anyone would be interested in doing. Some sweet tricks and tips for every day play in there too. Thanks for doing this and please continue to do so!

Just got to 21:41 in the run (43:26 in the vid) where you exploit the XP glitch. This really seems like the ideal opportunity to take the levels you need to exploit the mass haste, expeditious retreat and 3 levels of monk speed stacking, which results in character that exceeds the speed of even a level 39 monk. The formula is as follows:

The normal 1.5 * base speed cap to movement speed is overridden by monk speed (at monk 3) and stacking of speed bonuses occur as follows: Initial Speed = 1.00(base speed) + 0.1(monk bonus) = 1.10. Cast Mass Haste for 1.5 * Initial Speed + 0.1(monk bonus) = 1.5 * 1.1 + 0.1 = 1.75. Cast expeditious retreat for 1.5 * Mass Haste Speed + 0.1(monk bonus) = 1.5 * 1.75 + 0.1 = 2.725. That speed can outrun any natural monk (monk 39 moves at 2.30),

Examples of class builds that use this are below:



I'm sure you can tune it to use Wizard if you needed to, you would only need to be level 15 to have access to 1 cast of Mass Haste (12 Wiz/Sorc, 3 Monk), and refresh it as needed through quick resting.

I wasn't able to be faster with this trick, but maybe it could be useful in for example the beginning of HotU? I don't know, and right now frankly I haven't got the time to test it, however I'm still "archiving" the comment by quoting it here, in case it turns out useful.
Edit history:
ais523: 2017-01-11 06:32:14 pm
The problem with Nogaholic's method is that although it outruns a natural Monk, it doesn't outrun a natural Monk with permanent haste from items.

I don't think there's a way within the three official campaigns to get Expeditious or Mass Haste without repeatedly recasting them. Extend Spell would of course help, but the amount of extra spellcasting would notably cut into the gains, and you'd need to take way more than 3 Monk levels (with 3 monk levels, you have 2.725; with Robes of the Dark Moon plus 30 Monk levels, you hit 2.5, which is almost as good). Given that every eighth combat round has to be spent recasting your two spells (even Extended), we're talking 2.725 × ⅞ with the Monk 3 build, or around 2.38, which is slower than simply getting Haste from an item.

EDIT: I just remembered that Automatic Quicken is something that exists in the Diamond engine, but I don't know how it works. Writing this to make a note to myself to test it sometime, because if it's possible to cast the speedup spells for free, then the spellcaster build would win in terms of running around, and it's almost certainly better at combat too.
Edit history:
Mejcel: 2017-07-09 02:18:47 pm
Mejcel: 2017-07-09 02:17:06 pm
Mejcel: 2017-07-09 02:13:13 pm
Mejcel: 2017-07-09 02:10:34 pm
Mejcel: 2017-07-09 02:10:26 pm
Mejcel: 2017-07-09 02:10:12 pm
Mejcel: 2017-07-09 02:09:52 pm
Mejcel: 2017-07-09 02:09:31 pm
Mejcel: 2017-07-09 01:06:22 pm
I finally got around doing the improved HotU run. I got a crazy, almost perfect run, with a time of 26m43s. New stuff includes:

1. Start as a Monk -> manually level up to level 8 Monk, 3 Wizard -> auto level up to level 12 Monk, 3 Wizard.
2. In the Isle of the Maker, place a warp marker to cut some backtracking. I wonder why I didn't think of this before, it seemed so obvious.
3. Sell/buy back a page of full plates instead of Eyes of Petrification.
4. Using Improved Knockdown to save on Potions of Heal (and cut down time not in combat).
5. New extremely fast Mimic. Sometimes you get super lucky and Mimic's script forgets to eat the first gem. Also I figured out you can make do with just one gem if you pick it up and replace it quickly somewhere else.
6. Not getting the skill Discipline. This is super risky, probably doesn't save that much time and I don't really think it's worth it, but I did it anyway in this run. All the demons in Puzzler's Sepulcher for instance use Knockdown so you have to be somewhat lucky.

One thing to note is that in Chapter 3 I accidentally go to Quarry Office -> Hellbreath Tavern -> Quarry Office as if I would be fetching Eyes of Petrification again, while the optimal route would of course be Hellbreath Tavern -> Quarry Office. This cost me 6,5 seconds.

Might take a while to upload, but here's the link:

Edit history:
ais523: 2017-07-11 09:02:51 am
Glad to see that the new route is working out well!

I realised that many questions I had about the game mechanics could probably be answered simply by looking up guides written by other players (especially competitive players). One of the most notable facts that I learned, and didn't know before, is that Diamond does in fact have a speed cap, but the monk fast movement ability raises the speed cap considerably (from 1½ times normal, which is the speed cap in unpatched and for a non-Monk in Diamond, to a value equal to the monk speed + twice the normal speed). This means that any viable Diamond route requires at least three levels of Monk. The higher Monk speed cap seems to be impossible to hit, though, unless there's some method I'm missing, so it may as well not exist.

Quicken doesn't seem to be useful for the run; it has the same effect on spellcasting as Haste does, but Haste is easier to obtain. It's theoretically possible that it could be useful for casting Mass Haste, but even then, you'd probably have a regular Haste effect purely so that you'd be hasted while casting Mass Haste.
I can't really speedrun this game at the moment; my old Windows computer broke, and the one I can access at the moment runs Windows 10, which has the annoying habit of stealing focus from the game and/or becoming highly laggy at random moments. (I also seem to have lost my mouse, which isn't helping.) So instead, I decided to do something I've been meaning to do for a while: I wrote a memory-watch-like feature (via editing a script into the game that displays internal information), and decided to try to figure out how movement/clipping works (and in particular, what's up with the Docks clip). I'm focusing on Diamond here as it's what people (other than me) run.

Unfortunately, the results have left me more confused than before. It has confirmed some hypotheses I had, though. For example, it's now 100% definite that the character's hitbox and the character's model have independent positions from each other. When you do a door clip, the character model runs through the door, but the character's hitbox doesn't; it gets stuck on the door. Pressing W or S (and probably Q or E, but not A or D) pulls the hitbox onto the model's location, which is an almost unique effect in the game; in almost all other cases, the model gets pulled to the hitbox. There are a number of special cases to try to stop this being usable for clips, but they aren't comprehensive. (On unpatched, I've actually managed in the past to dissociate the model from the hitbox longer-term, i.e. "until I attempted to move". I haven't yet managed this over large distances on Diamond but I don't fully understand how it's done. I did manage a relatively small dissociation while randomly running around in the first room of Chapter 1, but don't know how I managed it, and that's the only time I've done it on Diamond.)

I think I understand the difference between familiar clipping on unpatched and Diamond. On both engines, when the familiar moves into the player's location, the game looks for a new position for the player. It does this by trying a sequence of positions in turn, and placing the player at the first valid location it finds. Each of the locations is determined via starting off with a "trial location", then picking a location near the trial location that's a valid square on solid ground (e.g. if the trial location is slightly inside a wall, the location it chooses will be on the standable ground next to the wall rather than inside the wall). I'm not 100% sure exactly how trial locations are changed to actual nearby locations. I do know, however, that the process can fail, causing a different trial location to be chosen.

Anyway, the main unpatched/Diamond difference is in the sequence of trial locations. These are always chosen by starting with the position of the object being pushed (the player in this case), and adding a vector based on the facing of the object pushing (in this case, the familiar). In unpatched, I don't have hard evidence of the exact pattern, but my experience with clipping strongly suggests that it's the following very simple pattern: 1 unit behind the pushee (in the direction the pusher is facing), 2 units behind the pushee, 3 units behind the pushee, and so on. The difference is that Diamond has also added other positions, e.g. 1 unit behind and 1 unit to the left, 1 unit behind and 1 unit to the right, 2 units behind and 2 units to the left, and the like. I haven't yet determined the exact pattern because it's incredibly hard to test, but on open ground, it always seems to be a whole number either orthogonally or diagonally (it's also possible that knight's-moves are in the list and I haven't observed them yet). This means that long-distance clips are going to be near-impossible in Diamond as there are so many more trial locations you'd have to block. There may well also be a distance cap; one other effect I occasionally observe is the player temporarily losing collision, which I expect is a consequence of no trial locations succeeding. One notable fact is that distance in this game appears to consider all 3 dimensions, so the Peninsula clip being impossible may well be a consequence of the fact that there's such a large y-coordinate change (making the clip appear to the game to be longer-distance than it seems from the map). Thin wall clips (such as the Callik clip) in Diamond appear to work by virtue of the fact that a point entirely on one side of the gate, and a point entirely on the other, are at least one units apart, so that the first trial location (1 unit behind) is unoccupied and the game doesn't even think about adjusting it to be on the same side of the wall (because it isn't in an invalid position); every time I've seen a thin wall clip succeed, it's moved the player directly to an unadjusted trial location. Note that one practically useful consequence of all this is that familiar clips will be more likely to work if the familiar is facing away from the wall you're clipping through; in Diamond, this is only a tendency, though, whereas in unpatched it's required.

The Docks clip outcome is only explainable by trial locations slightly off the bridge being considered valid (rather than invalid) and yet being adjusted to positions at the base of the bridge (rather than being adjusted to positions on the bridge or counting as invalid). I've seen other weird behaviour here in unpatched, but in Diamond, it seems to consistently choose the base of the bridge whenever it chooses a "weird" location.

Anyway, there's one thing I was previously mistaken about, and the memory watch has corrected me; hitbox positions aren't quantized in space at all, like I previously thought they were. For example, if you run along a wall at an angle, your hitbox becomes asymptotically close to the wall (i.e. the location of the hitbox moves continuously closer to the wall, but the rate at which it does so slows down so that it never actually reaches it, kind-of in the style of Zeno's Paradox). I still suspect there's some quantization going on, but it seems more likely to be quantization in *time*, i.e. the hitbox location doesn't update as fast as the character model position does. (This is a hard hypothesis to test due to limitations in the NWN scripting language, although maybe I'll think of a way.) Given that Neverwinter Nights was intended for network play, and computers were much less powerful at the time than they are today, it wouldn't surprise me if this was an attempt to minimize the amount of data that needed communicating between computers.

One other random fact I discovered, which doesn't seem useful but it can't hurt to document it: if you save and reload very near a door, you appear to move 1 unit perpendicularly away from the door.

Anyway, part of the hope with all this was to find new forms of movement (e.g. long-distance clips in Diamond, or warping to other areas on a map faster than running), but I suspect that at least with the basic mechanics we're currently aware of (mouse movement, keyboard movement, familiars), what we have right now is the optimum unless there are mistakes in the way the map tiles are designed (as there appears to be with the Aqueducts bridge). If there is some way to do a long-distance clip in Diamond, I suspect it involves covering more of the trial positions with monsters (either enemies or summons), but that seems unlikely (if not impossible) to be fruitful.