Posts for ais523

1 2
23 24 25
40 41
Posted: 1/5/2013 5:46 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
This run makes no sense, but in a good way. (Also, it can't be published in the Vault because it doesn't meet Vault rules. It seems to fit the Moon tier quite well, though.) Btw, how hexable is this game? If it hexes well enough, separate-cup TASes could eventually be combined into a full-game TAS.
Posted: 1/3/2013 7:57 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Yeah, SAT solvers can typically be beaten if you have more information about the problem space than they do. I've been wondering about keeping a cache of the fastest solutions for variousy-shaped solid color areas (this is one of the things I was planning to do when I split the puzzle into regions in my solver), but that isn't implemented yet.
Posted: 12/30/2012 5:45 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Where does the score counter max out? Playing it until that point, if it's not too high, would seem like a more reasonable goal. I fear this run would get repetitive if it went on much longer, though.
Posted: 12/28/2012 6:20 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
You could just store the tree of input, rather than the tree of savestates. The savestates can all be reconstructed from that. (This is how the "undo loadstate" command in nethack-tas-tools is implemented.) I do very much like the idea of a sort of side-commentary next to a TAS; it'd be similar in nature to an audio commentary, but allow you to show more details. It'd be a lot of work, but the end result would probably add to the entertainment.
Posted: 12/25/2012 1:33 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
You could probably farm the lives somewhere. Pretty much every Mario game has a spot where you can quickly farm infinite lives by getting an infinite combo off a shell and a staircase. "Warpless, avoiding coins" can be a pretty interesting goal in Mario games, though (e.g. I've had quite a lot of fun doing it in NSMB DS; the minimum is 7 coins, which can be accomplished with only 1 alternate exit, and perhaps even with 0). Requiring life farming seems a little out of place, though, unless it can be easily accomplished during any mandatory autoscrollers that might be in the run.
Posted: 12/21/2012 11:49 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
mathgrant wrote:
Hooray for thread necromancy, I guess. Regarding a no coin run, I tried that one time. I gave up on the first level. I'd love to see a no-coin run of as much of the game as possible, tool-assisted or otherwise.
WRT no-coin, you can get to 8-Tower1 with no coins, even warpless; I've done it on console. (Mini Mario is your friend.) And yes, the first level is surprisingly difficult that way. (I'm still very interested in whether 6-1 is possible coinless; that level can be skipped in a warpless by doing world 7 instead, but you can't do warpless no-alt-exit without 6-1, the best possible there is one alt exit, whether coinless or not. 6-1 is not obviously impossible, but I can't do it; the walljump section is too hard for me to avoid the coins in.)
Posted: 12/18/2012 2:30 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
I actually started route planning this game for a console run, but got fed up trying to optimize unlikely things as soon as the first mission. There are a lot of little time savers in the overworld. (The actual captures, of course, should be trivial to TAS, and also trivial to strategise; the strategy of "draw circles/triangles as fast as you can" is clearly the right one, and it seems unlikely that the time assists take to perform will save time in the actual captures. Even on Suicune. I have done no-assist runs of Suicune on console; on average, they take me around half an hour just for the one capture. A TAS would do it much faster, though.)
Posted: 12/15/2012 8:04 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Ever since the start of Fusion, Samus has had something that specifically makes her immune to X. I see no reason why it wouldn't work on something/one entirely different that also happens to be called X.
Posted: 12/12/2012 11:28 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
"The SM64 TAS team" would be reasonably descriptive.
Posted: 12/11/2012 5:34 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
We're likely to have a great screen shot possibility during the ascension run, and it won't even require any setup beyond what we need anyway. I think someone said they wanted us to fill a level with monsters, right? What about filling the most open level in the game with monsters in rainbow patterns?
Posted: 12/8/2012 11:16 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Anyone want an encode of the latest WIP? Here you go! This is a retimed version, with delays added and removed to make the run more watchable (we enter commands into the game at the rate of 1 per second; ttyrec players let you speed this up, slow this down, or frame advance). The encode is a "ttyrec" format file, that's typically used for recordings of text-based games. (Converting it to a video would make it much larger and also lower quality; ttyrecs are lossless, except for not tracking information about the font used.) It's of the Linux version of the run (running on a version hacked to sync with the DOS version), rather than the DOS version, so it'll look a little different to what you might be used to. All the keen NetHack players out there probably have ttyrec-playing software already, but for the TAS community, I recommend using Jettyplay, which I created specifically for the purpose of allowing the general public to play back ttyrec files. (It's unfinished, but entirely useable for this purpose.) BTW, we may have to redo the very end of this. The Archon's AI isn't cooperating on the ascension run, and we may need to force it to generate with an attack item; giving it more options on how to act ahead of time allows us to influence its AI ahead of time, because we already know what actions we'll be performing on the way up, and monster actions depend mostly on player actions.
Posted: 12/7/2012 6:29 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
HappyLee: with the new tiers, the site admins removed the rules on tier limits; you can have as many goals for one game as you like as long as they're all fun to watch.
Posted: 12/6/2012 5:28 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
We definitely hope so! 'Twould be a good day for a grand console verification (although adelikat and Nach tell me that those are meaningless on DOS), and a celebration of all things NetHack. (Especially as it's during one of the major NetHack tournaments.)
Posted: 12/6/2012 11:07 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Yeah, it works much better on games that have numerical restrictions on progression (like the star doors in Super Mario 64). It may be possible to define an uncontroversial definition even for other games too, though.
Posted: 12/6/2012 9:40 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
"No sequence breaking" is a legit category, much as it's causing me problems to get adelikat to admit its existence. There's already the occasional run in that category on the site already.
Posted: 12/5/2012 5:56 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Also check to see if changing the date on the DS changes the random seed. It does, for some games.
Posted: 12/4/2012 11:42 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
I think I found the bug, just testing at the moment. I'll let you know when it's fixed. 
Found connections at region edges:
+++++++.
+.......
.......+
+.......
.......+
.aaaaa?+
+!aaaaa.
.??????+

Best attempt before brute-forcing:
A?A?A?A?A?A?A?*?*?*
? ? ? ? ? ? ? |   ?
A?A?A?A?A?A?A-+ . *
? |               ?
*-+ . . . . . . . *
?                 ?
* . . . . . . . +-*
?                 ?
*-+ . . . . . . . *
?                 ?
* . . . . . . . +-*
?               | ?
* . +-B?B?B?B?B?B?B
?   | ? ? ? ? ?   ?
*-+-+ B?B?B?B?B . *
?     ? ? ? ? ?   ?
* . +-B?B?B?B?B?B?B
?   | ? ? ? ? ? ?
*?*?*?B?B?B?B?B?B B
This should be detected as unsolvable. The reason is easy: Take one starting point. Now solve the top part. If you now continue, then you come across the other endpoint and didn't solve the lower part. Now take a starting point, and assume you could solve the lower part. You can't get to the top part afterwards without coming across the other endpoint.
Yeah, I noticed this rule too, and is probably the most useful rule for solving that I haven't implemented yet. I need to work out how to express it in Lua. EDIT: Yeah, it was a simple typo in the solver that caused it to disregard solutions that used the border exactly once, that caused us to end up with those suboptimal solutions. Fixed version of the program is up at the same URL.
Posted: 12/3/2012 12:48 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
STBM wrote:
I was thinking about something... would it be possible to make a jumpless run ? It would require to get 70 stars (and so to actually find 70 stars which can be obtain without jumping), but it would be very interesting to watch...
It's been attempted, and I think there's a maximum-stars jumpless run in the thread somewhere. It can't get that far, though. (IIRC the first problem was no way to get into the pipe at the end of BitDW, but I might be wrong on that.)
Posted: 12/2/2012 9:45 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Yeah, about the worst you could do doing this on a console would be to trash the save file in such a way that the game crashed on load. The game's program is hardwired-in, what's running here is a separate program written into the game's RAM.
Posted: 12/1/2012 1:25 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
So, I had a go at writing a Polarium solver myself, and ended up with something massively sophisticated and overengineered. For anyone who's interested, here it is: http://nethack4.org/pastebin/polarium-solver.lua (Like other solvers here, I chose Lua for the potential to maybe implement into an emulator.) With luajit, this takes a couple of seconds for puzzle 100, and around 14 seconds for the 9 by 9 puzzle the latest TAS WIP starts with (after the "1"), to find the optimal solution in tiles and in frames. It's quite slow at puzzles with a lot of solutions, or a lot of red herrings; it works a little like I do when solving puzzles, although resorts to brute force at the end.
Posted: 11/28/2012 7:39 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
"Total control" makes sense as a category by itself, and it's the usual name I've heard for it.
Posted: 11/28/2012 4:57 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Thanks for the timing explanation. So it mostly but not entirely depends on the number of turns in the route; if I do write this solver (I have ideas about it, but am not really sure if they'll work), a route costs 7 for any line in the same direction, or 6 if it has length 1. (Plus a constant, but that doesn't matter when comparing routes.) Right? EDIT: Ah no. Dasaan's post explains it in enough detail for me, though. Thanks everyone!
Posted: 11/28/2012 4:27 PM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
bihan wrote:
marzojr wrote:
The rules say pretty much that -- you need a movie that starts from power on that creates the conditions necessary for the run, and the movie does not need to be optimized.
When would timing start?
Usually, you play the verification movie from the start until it creates the save you need to create the conditions necessary for run, use an in-game save at that point, and end the verification movie. You then start recording the actual run from power-on, but with that resulting save file. Get permission from a judge before doing this. Unapproved runs starting from a dirty save are often rejected for that reason.
Posted: 11/28/2012 11:46 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Ilari wrote:
Warp wrote:
Do I understand correctly that (analog) TVs actually adjust their own refresh rate to the TV signal that they receive (I'm assuming that within certain limits)?
Correct. It has to be done this way, since accurate enough timing sources would be hideously expensive (if available at all).
Yeah, when two things are communicating, it's nearly always best to have one of them drive a common clock. So either the TV or the console has to keep time for the other; the way it ended up working out is that the console does the timings. This also explains why you can damage a TV via turning up the refresh rate too high on a computer connected to it. Doing so is effectively overclocking the TV, and they have the same problems when overclocked as any other piece of hardware would.
Posted: 11/28/2012 11:19 AM
Quote
ais523
Editor, Experienced Forum User, Published Author, Player (44)
Joined: 7/11/2010
Posts: 1022
Can you explain what affects the time of a route? Is it the number of squares in the route, the number of turns it makes, or some combination? EDIT: And for anyone who doesn't know but wants to help, the rules of the game are that your line has to either go through all the black squares on a row but no white squares, or all the white squares on a row but no black squares. You can choose which color you're aiming for separately for each row.
1 2
23 24 25
40 41