Posts for ruadath

1 2
13 14 15
Posted: 2/5/2017 5:02 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
New WIP, completed through Chapter 2 Link to video
Posted: 2/4/2017 6:19 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Since I've now completed the Prologue, I've made this quick TAS of the first episode of the BS bonus chapters in just over 24 seconds. Link to video But it's probably not worth submitting this to the site (unless you guys think so?), though I probably will attempt a 999 score TAS and submit that. The other chapters are definitely much more involved and I probably will submit general TAS for those.
Posted: 2/2/2017 4:56 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Not really, unless you allow for the starting of a game from a premade bookmark on the first level. Even then, AI actions are controlled by a separate auxiliary RNG which cannot be controlled in real time, so you are certain to get a "desync" from a planned route fairly quickly.
Posted: 2/2/2017 3:21 AM
Post subject: Fire Emblem Awakening/Fates
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
I've disassembled and reconstructed the RNG for both of these games and will TAS them in a few years once Citra gains TAS capabilities.
Posted: 1/31/2017 4:32 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
The Shin Megami Tensei IV games Project X Zone (2) Fire Emblem: Fates Fire Emblem Echoes: Shadows of Valentia
Posted: 1/29/2017 4:43 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Thank you, it was quite a surprise getting accepted to Moons immediately for my first TAS... hopefully I'll have just as good luck on those that follow!
Posted: 1/29/2017 10:22 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Link to video After some more disassembly, some route planning, and about 350k resets for a good initial seed, here is the start of a Normal/Classic TAS of this game!
Posted: 1/28/2017 3:27 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Anyone have any thoughts about Zack and Wiki: Quest for Barbaros' Treasure? Also Battalion Wars 1 and 2 EDIT: Also Pandora's Tower
Posted: 1/26/2017 9:09 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
It ended up working out very well! The bot actually helped me discover a faster route that I didn't anticipate!
Posted: 1/25/2017 6:45 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Thank you so much! I'll give this a try and let you know how it goes.
Posted: 1/25/2017 11:06 AM
Post subject: Using a lua script to automate resets for seed (DeSMuMe)?
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Hey guys, wondering if you could help me with something. I'm currently TASing a game that fortunately does not have any clock-based lags. This means that given a possible desired outcome, I can write the movie file for it and find a seed that gives me the outcome later. So basically, what I'm asking is the following: I've got a move file of roughly 7000 frames, and I want to find an initial seed that gives me a particular result. To know if I got the "result," I just need to check and compare certain addresses in the memory with certain values. I've spent quite a few hours doing this manually (changing the clock time by 1 second, turboing through game movie, and then checking result), but am wondering if this can be automated entirely. Ideally, I would like such a script to dump the "good" datetimes (ones that give a seed that yields the result) into some text file/csv so I can go check those seeds later. For what it's worth, I'm fairly comfortable with programming and imagine I can probably achieve this with a .bat file or an autofire script, but I think things would be a lot cleaner (and maybe faster) if I could just run this through lua scripting. Any ideas? Code would be nice, but even just pointing me to a reference with the right lua commands (like for resetting the emulator and for editing the date in a dsm file would be useful).
Posted: 1/24/2017 7:58 AM
Post subject: Fire Emblem - New Mystery of the Emblem
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
So the next project I'm going to tackle is probably going to be TASing this game, since mechanically it is very similar to Shadow Dragon, and I have the RNG figured out for this game as well (though it is quite a bit more complicated here, so using a reverse lookup for the initial seeds isn't necessarily going to work...) Before I work on the game itself, however, it might be interesting to try individual TAS of each of the stand alone bonus chapters that the game provides. This does require having a save file that completed the first prologue chapters... and since I want to make sure I don't end up violating any rules this time, I'd like to know if it suffices just to create a verification movie that gives the desired SRAM?
Posted: 1/24/2017 2:02 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
OK, I have a movie that works now. I lost about 3 frames due to desyncs with the new clock seed, which I could probably get back by painstakingly searching for lags in the original run which were eliminated, but I'm too lazy to do that. Here is the submission.
Posted: 1/24/2017 1:13 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
So actually I think this might end up being OK. I found a different datetime in November 2071 that replicates the entire RNG stream except for the first one... and a quick run through didn't have any desync until frame ~16000 (roughly halfway through the movie), which I think is just caused by the fact that the arrow drawing mechanism in the game is based off the RTC and not the RNG... so I might have a fix that works up soon! EDIT: Or maybe not, the desync might be more serious here than I thought... we will have to see...
Posted: 1/24/2017 12:15 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Not sure who this question is addressed to, but if you're asking me, I'm certainly OK with it.
Posted: 1/23/2017 9:51 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
For what it's worth I checked what the error is. The revision is as dumb as changing a < sign to a <= sign in a for loop https://github.com/TASVideos/desmume/commit/1bce1f1811fa0a4e8802f5639b4ce6007c13a8b1
Posted: 1/23/2017 6:16 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
For what it's worth, I'm not interested in rerunning the game. As far as I'm concerned, this is a perfectly legitimate run only prevented by a bug in the official release of the emulator, and if you guys are not willing to accept that then I guess it just won't get published until 0.9.12 is released (if ever).
Posted: 1/23/2017 6:11 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Run is completed in 8:47.48! Link to video Due to TASVideos rules though, it will not be published on this website (the game starts running at a date time in December, which is not supported on the most recent official build of DeSMuMe). However, the movie does sync on an official build after a very small number of frames. As far as I am concerned, this is a perfectly legitimate run, so I will not be redoing this simply for the sake of a publication.
Posted: 1/23/2017 6:01 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
So should I cancel the submission then or what? The movie doesn't actually require interaction mid-run, you can run it from start to finish on (any of) the interim builds and it works. I just said that to show you that really the only thing going on here is the date time getting loaded in the first couple of frames. I figured this would a "special circumstance," since I imagine that if you run something like 5 or 10 frames (however long it takes for the emulator to load the date) that's all you need for the movie to sync on an official build.
Posted: 1/23/2017 5:43 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Right so what is happening is that the date parser in each of those builds is not reading the December date properly and is loading the game at Jan 01 2009 12:00:00 A.M. instead of Dec 26 2073 8:07:40 P.M. You need to load the first some number of frames (not sure what the lowest is, 31 definitely works) on one of the recent git revisions that fixes the parser so that the emulator loads the correct datetime/RNG. Then if you save state there and run the rest of the movie on the official build (or just continue to run on the git revision) it should work. Is this a problem/is there any way for me to fix this?
Posted: 1/23/2017 9:46 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
I just made an encode and am uploading to YouTube right now; there isn't any desync problem with this game on 0.9.10 or 0.9.11 I checked before I started making the movie. EDIT: I separately downloaded DeSMuMe (both versions), the ROM, and the submission onto a different laptop and it seemed to work fine. Not sure if this is sufficient? Also, considering that Vykan's movie, which was made on version 0.9.9 64bit syncs up with the git revision until frame ~30xxx (at which point a minor desync occurs), I have reason to believe that the game is reasonably compatible between different versions. Let me know if I screwed something up or how I can help though.
Posted: 1/23/2017 4:25 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Link to video New WIP with the route completely done since Chapter 4, saving over 20 seconds! Featuring better strategies, and more optimal menuing/execution, this video completes the game through Chapter 20 in just over 6:55, with Marth's Def high enough to survive Chapter 21 (which he had trouble doing in the last WIP)
Posted: 1/13/2017 8:48 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
For what it's worth, I've cracked the RNG formula (initial seed + next RN) for both this game and Path of Radiance (and I guess FE12 as well but that's not relevant). I'll probably be looking to tackle one of these as my next project once Shadow Dragon is completed and am looking for input as to which game to do first, which difficulty/version, RTA/Segmented etc.
Posted: 1/9/2017 10:57 PM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Link to video New WIP completed through Chapter 17 in a bit under 6:30. I used the clone glitch to pick up the third warp staff significantly more quickly than in the previous run. Expected time for completing the run is somewhere around 9:30, I think.
Posted: 1/9/2017 2:08 AM
Quote
ruadath
Experienced Forum User, Published Author, Active player (256)
Joined: 12/13/2016
Posts: 352
Thanks, I also wrote a script for New Mystery of the Emblem, and I might consider tackling that once this project is over (currently about halfway done)
1 2
13 14 15