1 2
12 13 14
18 19
Post subject: Blade Runner
Editor, Expert player (2479)
Joined: 4/8/2005
Posts: 1573
Location: Gone for a year, just for varietyyyyyyyyy!!
I know this is the future, because a casual conversation related to my hobby probes the existence of defining qualities that separate humans from robots.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
Personman wrote:
I more or less understand why this happened - we have some loose model of gameplay as by default only including interaction between the game and the player, and not the player and outside objects. For the bot, (some of) RAM is included in our perception of the bot's self, so it doesn't seem to violate the model, but for the human it does. Sorry for the random tangent!
Right, the difference here is that the RAM is part of the bot's "brain", but not the human's. The parallel breaks down because the equivalence here is "computer memorizes information" vs. "human tells computer to memorize information", which makes it clearer why the situation isn't quite symmetrical.
Post subject: Update to turnbyturn.txt!
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
So, we made progress on the NetHack run recently. Unfortunately, IO::Pty::HalfDuplex doesn't run on modern computers, which is making it hard/impossible for me to produce an encode of the run. Additionally, Gitorious is down, so the input file isn't even publicly available (I copied it to the TAS system so that it's backed up, but that's not of any use to anyone but dwangoAC or Ilari). However, I've updated the turnbyturn.txt writeup up to just before the last turn we TASed (I'm planning to redo the rest beyond there). Here you go: turnbyturn.txt
darkszero
He/Him
Joined: 7/12/2009
Posts: 181
Location: São Paulo, Brazil
This continues to impress me. What a novel way to steal from the shop! :)
Joined: 4/24/2012
Posts: 129
Location: Atlanta, GA
Just popping in to say I read the turn-by-turn today and was thoroughly impressed. I'm really excited to see what the rest of the run brings!
Joined: 5/13/2006
Posts: 283
The buildup to Bob was excellent. Keep it up!
<Zurreco> if so called professional players cant adapt to every playing field, theyre obviously not that great
Editor
Joined: 3/31/2010
Posts: 1466
Location: Not playing Puyo Tetris
Game? This is no game. It is a ground where bugs are abused and glitches exploited fully. It's great. Keep going and breaking it more.
When TAS does Quake 1, SDA will declare war. The Prince doth arrive he doth please.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
Wasn't done by me, but something very relevant to NetHack TASing: http://www.reddit.com/r/nethack/comments/3355z6/363_turn_16_second_set_seed_unnethack_win/ That's done on a variant of NetHack which has no T:2000 limit, and is likely a long way from the shortest possible UnNetHack win, but is an interesting start. (In particular, it uses only minimal luck manipulation, and that's used to survive.)
nht
Joined: 12/22/2010
Posts: 3
Thanks for the link ais523, I've made available the instructions to reproduce this run here http://nethackwiki.com/wiki/User:Nht/T363_Un_seed_314022_win_notes I also used some luck manipulation for stuff like getting decent rolls out of charging scrolls, wresting a wish (you only get one try in Un, though it's more likely) and getting djinn from lamps to grant a wish on the first #rub. I think 7 wall bumps was the most I did on any occasion. Though the random alchemy was the most frustrating, had a terrible run of luck getting water to become sickness first try. I mention in my reply to you on reddit some ideas for a v2 run on a different seed. The seed I'm thinking of is 53800, starting 2 steps from a wand of wishing(1:2) with another wand of wishing(1:2) on Geh:43. All Planes save Air can be sequence broke with death warps (via scrolls of flood), we start next to the portal on Air. Probably will be a monk to save a wish on the Eyes of the Overworld, then we can manipulate a branchport to skip traversing Quest:1. Will probably use the nethack-tas-tools for the luck manipulation. Is there a plan for where to move the repository when gitorious closes in a month?
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
Being git, we have local copies of the repository around. Our current tentative plan is just to make at least one of those local copies available readonly on one of our servers, and let people see what's going on like that. The branchport idea is to branchport into Home:1 in order to land nearer the quest leader than normal? I like it, that makes a lot of sense if you don't have to wait for a frame rule there. nethack-tas-tools has a few modifications to NetHack to be sync-stable with the DOS version (e.g. replacing the RNG with the RNG used on DOS, ignoring the passage of time, and the like). Presumably you'd be using the rerecording framework, rather than the modified NetHack? (Although the modified NetHack has a lot of useful tools too, like Santa mode.) I'd recommend talking to dwangoAC about it, anyway; he does most of the technical work on the rerecording framework.
nht
Joined: 12/22/2010
Posts: 3
Having considered the branchport to Quest:1 some more, I'm less certain it's worth the wish for the Eye, compared to doing a drowning deathwarp via eg. scroll of flood, or nearby water on the Tourist/Healer quest. Comparing the sequence of actions Branchport version
    * Wish for the Eye * Enter Quest:1 to get the option to branchport there * Leave Quest:1 * Branchport adjacent to Quest Leader
Deathwarp version
    * Pick up amulet of life saving (perhaps in 0 actions via autopickup) * Put on amulet of life saving * Enter Quest:1 * Read scroll of flood/jump over water (last action of the turn) * Polymorph from air elemental to non-flying form, safe_teleds adjacent to Quest Leader * Polymorph back to air elemental
(possibly 2 more actions if loadstones/inventory stashing are necessary) So it seems like a toss-up, especially considering the limited availability of wishes for magical items in Un. Intuitively I'd say the deathwarp version is preferable in all but the most optimal run. The wish is likely better spent on some other essential item that it otherwise takes more actions to obtain than on a single-use branchport saving 1-4 actions. Re: the nethack-tas-tools, I skimmed through the files. Currently I lack access to a machine capable of hardware-assisted virtualization, so can't use the qemu-kvm part of the rerecording framework. I toyed with using dmtcp for the application checkpointing part, criu/lxc is another option (lxc recently got checkpointing capability). This is all less general/portable than the qemu-kvm approach but will hopefully let me churn out another quick run before the lack of reseeding gets patched out for Junethack. The pexpect stuff will be useful, for manipulating teleports, wishes, polypiling, etc. The modified NetHack looks neat, though it's beyond my current abilities to port the changes over to UnNetHack. I don't think there's any pressing need to finish a close-to-optimal TAS of UnNetHack. The routes for NetHack variants with the frame rule removed are going to look fairly alike (as long as you're collecting the same McGuffins). One TAS would probably suffice across all such variants. Then, it's best to choose the variant that is easiest to work with or that has the most interesting differences.
Skilled player (1741)
Joined: 9/17/2009
Posts: 4981
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
Not exactly related, but the topic mentions a glitched any% in the first post. Has anyone managed to do that before?
Joined: 3/11/2008
Posts: 583
Location: USA
Personman wrote:
This is not meant as a correction/criticism, but merely an interesting philosophical observation that struck me reading the last two posts: it's kind of odd that it felt natural for you to distinguish between the actions "human takes notes" and "bot memorizes" in that way,
Not distinguishing these things is the idea of Extended Cognition (or Extended Mind), put forth in a paper in 1998 by Chalmers and Clark.
ais523 wrote:
"computer memorizes information" vs. "human tells computer to memorize information"
"computer stores information in RAM" vs "human stores information in notepad/computer's RAM"
However, I've updated the turnbyturn.txt writeup
Huzzah! "Santa mode"--Santascumming?
b -- a piece of coal.
Lovely use of those very particular random numbers; I'll not spoil it. (Though I am curious how many viable states there are for this kind of dodge of an Intended Rule; including the other possible items? Probably too large a state-space to search without a dedicated program, I expect, or possibly even with.) e: I was going to post "YARFU" Yet Another Request For Updates but you had, so it's slightly redundant. the typo "manipluate" remains in the text.
Joined: 1/29/2015
Posts: 21
I think I may have found some improvements for the finish. (Well, at least they were improvements on the plan for the finish as I last saw it. I cannot see what you have there now.) My first improvement is complicated and I am not sure whether it is feasible in practise. Here are the changes to make for this improvement. First, at the beginning of turn 2000, do not polymorph, and thus remain riding. The fifth and last action of turn 2000 (after level-teleporting to the bottom) should be spent running. Turn 2001 is the tricky part. First, I shall explain what needs to happen. We spend the first action ringing the Bell of Opening and the second beginning to read the Book of the Dead. The third action finishes it, and during the following monster action, we need to be hit by a quantum mechanic and teleported onto a level teleport trap, which is only possible after the first 200 random attempts of the teleportation fail, and enter Moloch's Sanctum. The fourth action should be spent running, and the High Priest of Moloch has to die during the subsequent monster action. The quantum mechanic and the High-Priest-of-Moloch-killer have to have started the turn as something else in order to have enough actions. This demands luck manipulation of the level's generation; we need a polymorph trap and a level teleport trap. If one or both of the traps is in a convenient location, the rest is easier, but because I do not know how likely this is, I shall assume that neither trap is in a convenient location. There are many things that need to be luck-manipulated here; the two chances we have to bump into walls do not seem to be enough, which is why I have come up with a new technique for luck manipulation for this. Here is the idea. In the middle of a monster turn, between two important random events, arrange for a monster to use a wand of fire and hit a pile of scrolls, which uses a random value to determine whether each of the scrolls is burnt up (see the function burn_floor_paper); the number of scrolls can be retroactively adjusted as necessary. Another option is to have a monster set off a land mine, causing random dispersal for each stack of objects there. My best idea for the High-Priest-of-Moloch-killer is to start with a fast tame air elemental carrying an extremely-highly-enchanted Frost Brand. The High Priest of Moloch starts with 25D8 hit points, 122.5 on average, so +80 should be more than enough. The advantage of this option is that little luck manipulation is needed for the polymorph, because many possibilities are acceptable. Putting everything together, here is the sequence of events in more detail. With the first action, do some regular luck manipulation and ring the Bell of Opening. There should be a fast quantum mechanic and two air elementals, one prepared as described above, near the polymorph trap. One air elemental steps onto the polymorph trap and turns into another quantum mechanic, and the first quantum mechanic sends it near the player. Most of the level should be filled with monsters or boulders to make the teleport more likely to go right. This monster must not disturb the player until it attacks -- this could be accomplished by making it peaceful and having conflict, or by making it invisible and not having monster detectio, or in some other way. With the second action, do some regular luck manipulation and begin to read the Book of the Dead. The second air elemental steps onto the polymorph trap and turns into an appropriate monster, and the quantum mechanic sends it near the level teleport trap. The third action is spent finishing the reading of the Book of the Dead. Now we may or may not need some midturn luck manipulation, and then the quantum mechanic sends the player onto the level teleport trap. We have 200 chances to hit one particular space out of 1659. The spaces vacated by monsters can have traps on them, should improving the chances slightly be necessary. Then we have some more midturn luck manipulation to get the landing position right (the actual level teleport happens after the monster turn finishes). Also, the pet wields the Frost Brand. The fourth action is spent running. The pet kills the High Priest of Moloch, and then the turn ends. At the beginning of the new turn, polymorph into whatever is needed next, drop off the steed, and land on the space on which the High Priest of Moloch died, grabbing the Amulet of Yendor by autopickup. (I am now not sure whether this one is possible, after trying it unsuccessfully.) My second improvement is to enter the tower's magic portal on a monster turn, making use of a quantum mechanic once again. This is not immediately possible, because the tower is contained in a special region that prevents teleportation in and out of it, and also prevents one from entering the level within it. However, if the initial position upon entering is next to the region and is occupied by a monster, there is a 1 in 3 chance of the player being relocated instead of the monster, and the region can be entered this way, and then a quantum mechanic can teleport the player onto the magic portal.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
n720 wrote:
I think I may have found some improvements for the finish. (Well, at least they were improvements on the plan for the finish as I last saw it. I cannot see what you have there now.) My first improvement is complicated and I am not sure whether it is feasible in practise. Here are the changes to make for this improvement. First, at the beginning of turn 2000, do not polymorph, and thus remain riding. The fifth and last action of turn 2000 (after level-teleporting to the bottom) should be spent running. Turn 2001 is the tricky part. First, I shall explain what needs to happen.
So in the new plan, turn 2000 goes: attack for alignment (1), gain quest entry (M1), levport (2), teleport and autopickup (M2), branchport (3), teleport (4), run into a wall (5). We might want to change the existing plan by removing the boulder fort and instead using the same pet (probably a leashed Archon) for multiple purposes.
We spend the first action ringing the Bell of Opening and the second beginning to read the Book of the Dead. The third action finishes it, and during the following monster action, we need to be hit by a quantum mechanic and teleported onto a level teleport trap, which is only possible after the first 200 random attempts of the teleportation fail, and enter Moloch's Sanctum.
Haha, beautiful. This is indeed something I hadn't considered. My problem here is how to get a quantum mechanic to act on M3. I'm pretty sure it's possible but it'd require a pretty difficult monster movement point action charging job.
The fourth action should be spent running, and the High Priest of Moloch has to die during the subsequent monster action.
…and here is where the trouble starts, as I can't think of a way for a monster to oneshot the HPoM with no inventory, and don't know of a way to charge a monster high enough to get an M4 action without using air-E form. This isn't completely out of the question, though. One thing I was wondering about was rearranging actions so that the monsters couldn't act immediately due to being off-level, which would throw off the monster action numbers in a way that's more benificial to us. Unfortunately, the two actions it takes to read the Book of the Dead have to be on the Vibrating Square level, and arriving on the VS level on action 1 doesn't help. Arriving on action 2 would help, but then we wouldn't have time to enter the Sanctum level. Possible exception: we change our subpixels so that the 5-action turn is 2001 rather than 2000, but then we're still trying to get a monster to act on M4 (which comes after our own action 5).
The quantum mechanic and the High-Priest-of-Moloch-killer have to have started the turn as something else in order to have enough actions. This demands luck manipulation of the level's generation; we need a polymorph trap and a level teleport trap. If one or both of the traps is in a convenient location, the rest is easier, but because I do not know how likely this is, I shall assume that neither trap is in a convenient location.
I believe I can get the polytrap into a convenient location, at least. The levport trap will probably have to be random, but it doesn't really matter where it is anyway.
There are many things that need to be luck-manipulated here; the two chances we have to bump into walls do not seem to be enough, which is why I have come up with a new technique for luck manipulation for this. Here is the idea. In the middle of a monster turn, between two important random events, arrange for a monster to use a wand of fire and hit a pile of scrolls, which uses a random value to determine whether each of the scrolls is burnt up (see the function burn_floor_paper); the number of scrolls can be retroactively adjusted as necessary. Another option is to have a monster set off a land mine, causing random dispersal for each stack of objects there.
I was thinking "why not use pet-attack-cancel", but we're in Gehennom and so have tons of walls anyway. The issue here is luck manipulation on monster turns, right? I believe the simplest method would be to place monsters that are manipulated into AI branch 1 (which is quite easy: there are tons of things that force monsters into AI branch 1, including not being adjacent to the player), so that they consume random numbers in their AI code. We'll have to place a lot of monsters anyway in order to get a 200 random number failure with a chance that's meaningfully possible to manipulate.
My best idea for the High-Priest-of-Moloch-killer is to start with a fast tame air elemental carrying an extremely-highly-enchanted Frost Brand. The High Priest of Moloch starts with 25D8 hit points, 122.5 on average, so +80 should be more than enough. The advantage of this option is that little luck manipulation is needed for the polymorph, because many possibilities are acceptable.
The problem is that air elementals won't wield weapons, and monsters spend an action to wield a weapon (and only if they intend to attack the player). All that said, I think the turns immediately after T:2000 (and thus probably the whole asc run) will have to be done under conflict, meaning that we can (probably?) get a tame monster to wield a weapon in order to attack us.
Putting everything together, here is the sequence of events in more detail. With the first action, do some regular luck manipulation and ring the Bell of Opening. There should be a fast quantum mechanic and two air elementals, one prepared as described above, near the polymorph trap. One air elemental steps onto the polymorph trap and turns into another quantum mechanic, and the first quantum mechanic sends it near the player. Most of the level should be filled with monsters or boulders to make the teleport more likely to go right. This monster must not disturb the player until it attacks -- this could be accomplished by making it peaceful and having conflict, or by making it invisible and not having monster detectio, or in some other way. With the second action, do some regular luck manipulation and begin to read the Book of the Dead. The second air elemental steps onto the polymorph trap and turns into an appropriate monster, and the quantum mechanic sends it near the level teleport trap. The third action is spent finishing the reading of the Book of the Dead.
Now we may or may not need some midturn luck manipulation, and then the quantum mechanic sends the player onto the level teleport trap. We have 200 chances to hit one particular space out of 1659. The spaces vacated by monsters can have traps on them, should improving the chances slightly be necessary. Then we have some more midturn luck manipulation to get the landing position right (the actual level teleport happens after the monster turn finishes). Also, the pet wields the Frost Brand.
The problem being that we need to hit 200 wall or monster squares first. The only sensible thing I can think of to do here is to work out exactly what RNG seed we'll be using in advance, and then generate a monster on every non-wall square that would be randomly hit.
The fourth action is spent running. The pet kills the High Priest of Moloch, and then the turn ends. At the beginning of the new turn, polymorph into whatever is needed next, drop off the steed, and land on the space on which the High Priest of Moloch died, grabbing the Amulet of Yendor by autopickup.
So that's why you stayed on the steed all this time, clever. I was aware you could get a free movement like this, but hadn't thought of using it here. Additionally, we have no walls to bump into to end the running at this point, but falling off the steed should end it. I take it all this is just to save one action? Both this plan and the old one have their first post-Amulet action at the start of turn 2002. The difference is that with this plan we get five actions on turn 2002: we have 11 movement points (best case) after the last action of turn 1999, thus at speed 53, we have 11+53 = 64 at the start of turn 2000, 4 at the end of turn 2000, 57 at the start of turn 2001, 9 at the end of turn 2001, 62 at the start of turn 2002, giving us another 5-action turn (and leaving us at a "safe", i.e. doesn't disrupt the plan and doesn't cause any glitches, subpixel value of 2).
(I am now not sure whether this one is possible, after trying it unsuccessfully.) My second improvement is to enter the tower's magic portal on a monster turn, making use of a quantum mechanic once again. This is not immediately possible, because the tower is contained in a special region that prevents teleportation in and out of it, and also prevents one from entering the level within it. However, if the initial position upon entering is next to the region and is occupied by a monster, there is a 1 in 3 chance of the player being relocated instead of the monster, and the region can be entered this way, and then a quantum mechanic can teleport the player onto the magic portal.
The reason why this didn't work for you is that there's a range limit on doing that. You can only move one square, and there's an undiggable and unphaseable wall around the edge of the teleport region. So this won't fly. In all, I'm not sure if we can get a 2014-turn ascension like this. We reach Earth one action sooner, which is big, but there's no obvious way to rejig that into an entire turn of savings. I know there were various other plans we had for the Planes that saved time but not a whole turn; I'll have to look into them in order to see if they can be adapted to save a turn given the extra action saved earlier. EDIT: New hypothetical plan for Rodney's on the way up. We enter the tower on the monster turn after our third action of turn 2002, as before (enextoing into the magic portal to arrive on the level). However, instead of just arriving there normally, a ton of (precharged) monster actions happen: we have the Amulet stolen by a covetous monster, get engulfed, then a tame monster that we'd left there kills the engulfing monster (we could have left it on low hitpoints), we fall out back onto the portal and end up back on fakewiz1 (G:43). We now have one action left (original plan) or two (n720's plan). Then we levport up onto wizard3 (G:38), the covetous monster teleports after us (leaving the tower), something kills it (tame or conflicted monster), then we get Q-teleported onto its corpse and pick up the Amulet. We also need to hit a polytrap somewhere along the way in order to turn back into an air-E (so as in the old plan, all our MR-providing items need to be stolen). Sadly, this is no faster than the original plan (although it's more stylish because it doesn't abuse enexto to leave the tower). I was hoping to save an action either like this, or by throwing in a 200 teleport fail and levport trap somewhere to save a manual teleport, but the problem is that we can't enexto twice on the same monster action and (due to teleport regions) can't be horizontally teleported outside the fake wizard's tower (which doesn't generate levport traps). Actually, I just realized that this doesn't work because wizard3 is no-teleport (and thus we couldn't pick up the Amulet without losing time). Anyone have any ideas to fix it? EDIT to EDIT: if the covetous monster is also a follower, is it possible to get it to ship the Amulet up to a teleport-allowed level for us via following, without losing time? This depends on the exact details of the mysterious force rules, which I don't know offhand.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
OK, time for a list of known and potential improvements to the current plan. Might not save a whole turn yet, but it's getting close. For reference, the old plan can be viewed here. Amulet of life saving shenanigans The first thing to note about the route is that the turn boundaries on Air and Astral are fixed in place. We can't do anything on Air without the turn boundary filling it with monsters, and thus in order to save a turn, the turn has to be saved entirely before the turn boundary in question ("2013:^" in the linked notes), or entirely afterwards. However, we can fool this restriction somewhat. There's a glitch with amulets of life saving that means that certain commands can be caused to take no time if you die and are lifesaved during the command. (Specifically, this works only on commands that have no special handling for numeric prefixes.) This means that we can save one action before turn 2013 in the current route (2012 in a hypothetical 2014-turn route) via using the amulet of life saving action cancel glitch, but at the cost of one action after that turn (to put a new amulet of life saving on for Water). This is a good trade, because it lets us consolidate savings in one place rather than waste them to frame rules. The action that we save before turn 2013 is actually one of the very first. On our second action of turn 2000, we level teleport down to the Dark One's lair. This action may well happen in natural form (see n720's post above, or my summary below), meaning that we can be vulnerable to HP death. One of our options for level teleportation is to invoke the Orb of Fate, which does a small amount of HP damage to us. If we're on critically low health at the time, we can die, lifesave, and cancel the action (meaning that we get six actions on turn 2000, which is probably a new record). To pull this off, we also have to save an action somewhere on Air, Fire or Water. I'm currently unsure on how to do this, but there are plenty of potential suggestions (a promising one is to get pushed out of the water bubble we're in on the turn boundary after doing Air, thus saving us the first action of that turn; another possibility is to use air elemental polyform + the Eyes of the Overworld, with the disadvantage that it prevents using polymorph traps or level teleport traps on ourself earlier in the ascension run). 200 attempt teleport failure People who have experimented with NetHack in debug mode generally know you can't teleport onto a trap. I knew that too, which is why I didn't factor it into my plans. As n720 points out, though, that isn't technically correct; if the game picks 200 sets of coordinates for a random teleport and they're all ineligible, the game starts looking at traps as options too. This means that we can level teleport on the monster turn (something that isn't normally possible) via use of a quantum mechanic teleporting us onto a level teleportation trap (likewise, we can polymorph on the monster turn via the same mechanism). We can arrange for this to happen anywhere in Gehennom proper that isn't no-teleport, before we get the Amulet; the problem is that we can't have any MRtifacts in inventory at the time (the main problem artifacts being the Eyes of the Overworld and the Eye of the Aethiopica). This doesn't directly save any of the actions in the plan I linked above. However, as n720 mentions, it allows us to rearrange actions on turns 2000 and 2001 to get the same result in natural-form-riding as it would in air elemental form: 2 actions lost to: the farmove actions needed to keep natural-form-riding happening at top speed. 1 action gained by: quantum mechanic teleport + 200 attempt teleport failure to enter the Sanctum; there's some subtlety here because monsters don't move after you enter a level on a monster turn, which n720's plan solves via doing the gallop farmove on that turn; if we combine this with the action saved to a "oLS then the timings don't work out correctly for that any more but there might nonetheless be some way around it via rearranging actions or saving some other action with a "oLS. 1 action gained by: falling off our steed onto the Amulet of Yendor at turn boundary, rather than having to step onto it manually. So a wash, right? Actually no; air elementals are speed 48, a galloping unicorn is speed 53 (but is slower for general usage because you need to spend the last action of each turn moving, which is normally an inefficient use of time). We can spend three turn boundaries galloping this way, when with the old plan we'd only spend one. Perfect movement subpixels at turn 2000 are 11. 11 + 53 + 48 + 48 = 160 = 12*13 + 4, i.e. we have 13 actions total on turns 2000, 2001 and 2002 in the old plan. 11 + 53 + 53 + 53 = 170 = 12*14 + 2, i.e. we have 14 actions total on turns 2000, 2001 and 2002 in the new plan. So the upshot of all this (besides making the TAS way harder to make because the monster actions happen at some of the most awkward possible times that are theoretically possible to hit) is that we can gain one action via this particular string of coincidences, assuming that there's some fix for the problematic details. Portal + level teleport to sequence break the Wizard's Tower This was actually the first sequence break I discovered in NetHack, then I stopped using it in favour of enexto. It's a lot more stylish, and possibly easier to set up. But does it save time? The original plan involved dropping the Amulet, then using a wand of teleport to teleport it out of the tower, but that's rather slow. There's an alternative, though; we can have a covetous monster steal it from us and then covetous-teleport out of the tower. (The only potential covetous monsters that will do this are doppelgangers of quest nemeses, with only a 5% chance of succeeding even if they use the attack in question, but we've never let the absurdity of something like that stop us before.) If we enter the portal to wizard3 on our own action, then we can get engulfed and unengulfed on the following monster action, dropping us back on the same portal and sending us back to fakewiz1. The previous plan involved enextoing onto the portal (a monster-turn thing), but we could instead use punishment dragging to cover the short distance we need (note that we haven't been an air-E at any point before this turn, so we could set up a punishment ball before turn 2000 and just leave it in inventory, weighed up by the antigold). Note that we can't use jumping because that costs a whole turn. This means that in both the old and new plans, the Amulet ends up on wizard3 the turn after we enter fakewiz1. The difference is where we end up (wizard3 in the old plan, fakewiz1 in the new plan). The followup in the old plan is to enexto out of the tower onto a polytrap and turn into air-E form (we need to get out of p/t form). The corresponding followup in the new plan is to levport to wizard3, then on the monster turn have the covetous monster teleport over to us, and get killed by another monster, dropping the Amulet next to us. Unfortunately, even if the Amulet is dropped onto a polytrap (the optimal case from our point of view, as we want to become an air-E during this turn), we then probably have to spend an action picking it up, making this plan one action slower despite everything (sadly, we can't use the monster turn for help because wizard3 is no-teleport, and monsters with the Amulet never follow across levels). It is probably possible to "oLS-cancel the action on which we pick up the Amulet (say, via autopickup of a cockatrice corpse without gloves), meaning that this option may be competitive if we need to find some other opportunity for "oLS-cancelling than the one on turn 2000. I'm very sad that I can't find a way to make it faster, though. Another possibility is to somehow get the covetous monster in question to leave the tower on the monster turn during the first visit, after stealing the Amulet, so that it can die there. The only potential for this involves the 24 subpixel movement alternation glitch, and somehow levporting a nymph who has stolen the Eye of the Aethiopica to the level. Perhaps not impossible, but it'll depend heavily on where the turn boundaries are, maybe requiring "oLS-cancelling or the like to move them around. Still another possibility involves riding for five turns rather than three. The problem is that riding for four turns is just outright cost-ineffective, and by turn 2004 (the fifth turn) we should be chain-quaffing "oLS. Conclusion Even if these strategies can somehow be made to work, we're still only saving three actions before turn 2013. To get a 2014-turn run, we need to save four. Thus, we still need to find extra savings somewhere, or else 2015 is still the theoretical limit. 2014 is not looking as insurmountable an obstacle as it did a few weeks ago, though. Anyone got any more ideas?
Joined: 6/6/2004
Posts: 223
ais523 wrote:
The problem is that riding for four turns is just outright cost-ineffective, and by turn 2004 (the fifth turn) we should be chain-quaffing "oLS.
"Molten Amulets: Xorns' Favorite Energy Drink! Now available at Sablja's Shop!"
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
Major new discovery that will make the route a lot faster. (Thanks to Corwinoia and FIQ for asking the right questions.) <Corwinoia> can you drag a covetous monster with the amulet with you through a level port? <ais523> Corwinoia: no <ais523> monsters with the Amulet don't follow <ais523> nice try though ;-) [...] <Corwinoia> ais523: a pet carrying it? <ais523> Corwinoia: still won't follow through levports <FIQ> leashed pets? <Corwinoia> branch port? <ais523> FIQ: nope! (dog.c line 518, had to check to make sure) <ais523> err, 518 onwards <ais523> there's even a unique message for that case [...] <Corwinoia> wait.. 529. If you put it on your mounted ki-rin? [...] <ais523> leaving a steed behind without dismounting it would be terrible, thus the check has to be somewhere else <ais523> but I can't find it <ais523> so it might have been left out altogether <ais523> oh, the check is apply.c line 533, but that seems like the wrong place so it might not actually work <ais523> ah yes, teleport.c line 688 applies it <ais523> so the level teleport is just blocked outright if your steed has the Amulet <ais523> and next_to_u is /really/ badly named <Corwinoia> where is next_to_u called from? <ais523> teleportation functions <ais523> its purpose is mostly to ensure that pets follow you through a teleport, but it can also veto a teleport <ais523> which it apparently does if your steed has the Amulet <Corwinoia> hmm <FIQ> ais523, what about branchporting? <ais523> NetHack has something of a history of just putting code in random places that does much the right thing <FIQ> they're different, for some reason <ais523> FIQ: oh wow, the next_to_u call seems to be missing <ais523> if you could let your steed pick up the Amulet, then branchport to Sokoban… <ais523> this really needs testing <FIQ> it would actually work? * FIQ wizmode <ais523> I might be missing something <ais523> definite wizmode time, I'll race you <Corwinoia> ais523: it's not there for branch porting <Corwinoia> or I don't see it <ais523> Corwinoia: I don't see it either Turns out that this works: if you're riding a steed that has the Amulet, most forms of level teleportation won't work, but branchport will because the `next_to_u` check is missing. This has major implications for the route because we no longer need to do the c!oGL run through Gehennom, and should save many turns. The bad news: because this lets us sequence break sanctum→sokoban4 rather than fakewiz1→wizard3, the dungeon layout that we need for an optimal sequence break is different. The one that we have in the current TAS work in progress is one action away from optimal. I don't yet know whether this will lose a whole turn or not; if it doesn't, it doesn't matter. The other bad news: the current TAS work in progress gets a bunch of potions which we no longer have any use for, which looks like a sloppy planning mistake in its own right. (In other news, I found a different way to beat 2015 but it's not as fast as this one, and would have required some tedious grinding of the 'riding' skill: if you let a monster go up through the dungeon via chain c!oGL quaffing rather than doing it yourself, it takes the same amount of time in player actions due to some dubious coding in allmain.c, but you have much more freedom in what those actions are. This means that you can maintain a gallop all the way up and do it at speed 53 rather than 48.) EDIT: FIQ just realised that it's faster to branchport to the top Mines filler level, which is possible if it's the only Mines level you've visited. This is two actions away from its optimal position in the current TAS, which is very bad news; it means that we have a pretty high chance of needing to restart in order to manipulate the dungeon structure to give us the fastest possible ascension run.
Joined: 5/13/2006
Posts: 283
Oh man that's a huge discovery. Shame all that work might need to go away for it though.
<Zurreco> if so called professional players cant adapt to every playing field, theyre obviously not that great
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
Update: stenno just tested this in an actual game (no TAS tools or debug mode or whatever) to make sure that we hadn't missed something. It definitely works. As for what to do about the current TAS WIP, it's a bit awkward. Most of the things we did will go much faster the second time round, assuming that we do redo the run (at this point, we'll probably have to); we have the tools in place for them now, and the strategies worked out already, even if we'll need to do it on different levels. dwangoAC wants to finish the current TAS first as a testrun, but we haven't really set up for sort of the ascension run that becomes possible with this glitch. (First thing to do, though, is to retime everything post-T:2000 to see just how fast the new route is.)
darkszero
He/Him
Joined: 7/12/2009
Posts: 181
Location: São Paulo, Brazil
More hype! Really loved the chain of guesses from a covetous monster, to a pet to your mounted ki-rin.
Joined: 1/29/2015
Posts: 21
ais523 wrote:
(I am now not sure whether this one is possible, after trying it unsuccessfully.) My second improvement is to enter the tower's magic portal on a monster turn, making use of a quantum mechanic once again. This is not immediately possible, because the tower is contained in a special region that prevents teleportation in and out of it, and also prevents one from entering the level within it. However, if the initial position upon entering is next to the region and is occupied by a monster, there is a 1 in 3 chance of the player being relocated instead of the monster, and the region can be entered this way, and then a quantum mechanic can teleport the player onto the magic portal.
The reason why this didn't work for you is that there's a range limit on doing that. You can only move one square, and there's an undiggable and unphaseable wall around the edge of the teleport region. So this won't fly.
I was referring not to the main tower, but to the small one that leads to it: that one is enclosed only by water and ordinary walls. It probably doesn't matter now that you have a way to bypass that. Here is something else I found that may be useful. Earlier, hurtling by throwing things was brought up as a way to pass the plane of Air, but in the end the limit of 9 made it inferior to other approaches. But there is another way to hurtle there: kicking a monster. In this case, the distance is determined by a ratio of weights and there is no limit on it.
Moderator, Senior Ambassador, Experienced player (907)
Joined: 9/14/2008
Posts: 1014
n720 wrote:
In this case, the distance is determined by a ratio of weights and there is no limit on it.
Hmm... As in, the total player weight, including inventory? With all that antigold we're not particularly heavy. :) I'm not sure if this will work or not but it's definitely an interesting thing to discuss with ais523.
I was laid off in May 2023 and became too ill to work this year and could use support via Patreon or onetime donations as work on TASBot Re: and TASBot HD is stalled. I'm dwangoAC, TASVideos Senior Ambassador and BDFL of the TASBot community; when healthy, I post TAS content on YouTube.com/dwangoAC based on livestreams from Twitch.tv/dwangoAC.
Editor, Player (44)
Joined: 7/11/2010
Posts: 1029
New tentative route for the quest and ascension run (with explanations and the like) is here. It comes to 2006 turns. Not final yet; there may be mistakes, and there's just enough inefficiency there that it seems as though 2005 might potentially be possible (although I'm not as of yet sure how to do it).
Patashu
He/Him
Joined: 10/2/2005
Posts: 4043
2000:M1 200 teleport fail into a hole trap, and fall down to the
        Vibrating Square level, autopicking up the lit Candelabrum
        (which can't have been in inventory at this point or it would
        have been extinguished when we fell ino the water). Because
        this is the last action of the turn, this hypercharges the
        monsters on the destination level.
What water is being referred to here?
2002:58 Zero-time artiblast depolymorph. (We could just polymorph at
        turn boundary, but the previous turn is so low-probability
        that manipulating a polymorph at 2002:^ would be out of the
        range of plausible luck.)
Does this require appearing on an artifact and autopickuping it, or are there actions that blast you with an artifact from your inventory and take no time?
My Chiptune music, made in Famitracker: http://soundcloud.com/patashu My twitch. I stream mostly shmups & rhythm games http://twitch.tv/patashu My youtube, again shmups and rhythm games and misc stuff: http://youtube.com/user/patashu
1 2
12 13 14
18 19