Post subject: Fire Emblem - Shadow Dragon
Rolanmen1
He/Him
Experienced player (751)
Joined: 2/20/2009
Posts: 569
Location: Dominican Republic
I made this WIP that finishes the Prologue. It used the default clock value when you start recording a movie. And by far, i got some good luck. Prologue 1: I managed to get a crit on the boss plus a good level up, this crit saved me to attack in Player Phase. Prologue 3: Here i had to give Marth a Vulnerary and since the Jeigan is the unit most near with one i traded it with him, i also traded his Iron Sword to him (not sure if this is a good or a bad decition). I also had to move one of the Cavaliers because the AI would move one Soldier in Marths way. Avoiding this saves one turn. Prologue 4: I had to move the Jeigan along with Marth so the Jeigan kills Gordon, otherwise, Marth would have to get a crit in Gordon and he would get a level up during Player Phase wich is something really bad. About a H5 TAS. I don't see this comming anytime soon, unless someone does an intense research on the RNG System. A way to do a TAS of this is by suspending a soft resetting in different seconds to PROBABLY get a good set of RNs. Please, give me your opinions about this WIP. EDIT: Adding some info i forgot to post before: This was done on DeSmuME 0.9.8 and with Disabled Advance Bus-Level Timing. And here's a small encode.
Editor, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Out of curiosity which difficulty mode did you choose? I know one of the TASvideos guidelines is to choose the hardest difficulty possible from a clean SRAM, but with this game the hardest difficulty doesn't work very well with the "beef the main character as much as possible and blindly attack anything and everything" strategy.
Homepage ☣ Retired
Rolanmen1
He/Him
Experienced player (751)
Joined: 2/20/2009
Posts: 569
Location: Dominican Republic
Comicalflop wrote:
Out of curiosity which difficulty mode did you choose? I know one of the TASvideos guidelines is to choose the hardest difficulty possible from a clean SRAM, but with this game the hardest difficulty doesn't work very well with the "beef the main character as much as possible and blindly attack anything and everything" strategy.
A H5 (Hardest Difficulty) TAS of this game is very much impossible. I mean, your characters are weak in comparison with your enemies, and since you can't manipulate in this game that well, you have high probabilities to end up having bad characters anyway. If we can find a way to burn RNs in a quick fashion like the GBA Fire Emblem does, i would so much go for that, otherwise, i would go for Normal Mode.
Joined: 5/9/2005
Posts: 748
How can a TAS be impossible? Are you saying picking Hard5 makes the game un-winable?
Rolanmen1
He/Him
Experienced player (751)
Joined: 2/20/2009
Posts: 569
Location: Dominican Republic
Paused wrote:
How can a TAS be impossible? Are you saying picking Hard5 makes the game un-winable?
Maybe i should pick my words better. It's winable, definitely. But you can't do manipulation during game (so far), with this in mind, to get a good time, you have to be really, REALLY lucky. Or just make a slow run all kind of unnecesary stuff.
Joined: 5/30/2006
Posts: 39
While I like that you're working on this game, I'm not sure that a TAS is a good idea due to how the RNG works. It just seems like you'll be at a massive disadvantage compared to a standard segmented run and I don't see it ending well. I see Marth is already down 2 strength and 1 defense, so that's concerning too. My memory is fuzzy, but I'm fairly sure there was something important about gaining strength almost every level early on. Have you made sure to check boss stats to find out if you're still on track? As for Hard 5, I figured a TAS would just use the "Warp Sheeda, kill boss, Warp Marth, and win" strategy for a bunch of chapters. Trying to imitate a Normal run's solo Marth plan would only end in tears anyway.
Former player
Joined: 5/29/2006
Posts: 200
I agree that this is a poor game choice until and unless a way to burn RNs is found. When/if one is discovered the game will probably end up being sub-20 minutes with enemy phase skipping and warp staff abuse.
Joined: 7/22/2010
Posts: 53
The guidelines say the hardest difficulty is preferred unless the only difference is the enemy/boss hit points, in which case choose the fastest difficulty. That's basically the gimmick behind all the Shadow Dragon hard modes, each one just has progressively more overpowered enemies. So Normal mode isn't really a bad choice... Except that hard mode skips the prologue chapters. So I wonder, is Hard 1 maybe the fastest difficulty choice? Or does "Uses easiest hardest difficulty" just sound dumb? Either way, I agree w/ the above that this game really needs its RNG cracked to be a good game choice.
ALAKTORN
He/Him
Player (99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
My God, do we not have a proper topic for this series? I’ve seen Toothache (I think that’s his name) post TASes on YT, does he frequent here? I want to get information on FE’s RNG.
Active player (276)
Joined: 4/30/2009
Posts: 791
From my understanding the RNG in this game is very fixed, even more so than the GBA games, because you can advance the RNG to where you want. It is understood if you change the internal clock you can change the RNG, but that's the only known way to do so atm, would be good to explore to find a way that doesn't involve such a slow method.
Player (98)
Joined: 3/20/2008
Posts: 466
Location: Montreal, Quebec, Canada
I recently completed a single-segment (RTA) run of the game Link to video One thing that should be noted is that you really don't need much luck with the RNG to clear the game this fast. As long as you can ensure Marth is def blessed early on (which I managed with a clock state of 12:00:10 AM, January 2009 on cleared SRAM), Marth will easily be able to solo the game with an 8 mt rapier forge. Obviously it's more ideal to crit every boss in the game and avoid the C4 armory, but without a powerful means by which to manipulate the RNG, that is just a concession we'll probably have to make. It's not as though the RNG isn't manipulable at all. The RNG changes as a function of time whenever there's dialogue on the screen, and I believe it changes dramatically faster if you don't set text speed to max. There are certain cutscenes you can skip a little slower than usual to get different RNG, you can change the order in which units attack, attack with different weapons, etc. I was actually planning to whip up a quick TAS that basically imitates my RTA run and foregoes any deliberate RNG manipulation. Probably wouldn't take more than a day or two to perform. If anyone wants to help me crack the game's RNG, it is located at addresses 02196E08 to 02196E0B as four one-bit signed values. This might also be of some help. I can track as many initial RNG states for different DS clock times as necessary. It's also possible to burn 1 RN at a time by simply waiting in an armory, which provides plenty of raw data for finding an RNG formula. I just don't know how to extrapolate a formula from even a large sample size of data, perhaps someone with more expertise can help me out on that end. As a side-note, FE12's RNG is time-based and varies every frame. Only half the game can be warpskipped, and the game contains some very interesting staves, particularly the rescue, thief and again staff. I haven't even been able to determine where the game's RNG addresses lie, but if someone were interested in helping me with that, I would definitely TAS that game in a heartbeat.
Active player (255)
Joined: 12/13/2016
Posts: 352
function nextrng(r1)
        local seed = {0x18EE6547,0x8C7732AF, 0x463B9957, 0xA31DCCA7}
        return AND(XOR(r1, SHIFT(r1, -3), SHIFT(r1, -4), SHIFT(r1, 3), SHIFT(r1, 4), seed[r1%4+1]),0xFFFFFFFF)
end
 
function rngsim(n)
        local rngbase=0x02196E08
        local result = { memory.readword(rngbase)+memory.readword(rngbase+2)*65536 }
        for i = 2, n do
                result[i] = nextrng(result[i-1])
        end
        return result
end
 
while true do
        local nsim = 47
        rngs = rngsim(503)
        for i = 1, nsim do
                gui.text(236, 8*(i-1)-190, string.format("%3d", (rngs[i]%2147483648)%100))
        end
        gui.text(160,-190,"Previous RN:")
        emu.frameadvance()
 
    end
lua script for the rng in this game is now available; clock seeding mechanism still needs to be determined EDIT: Clock function is monic and linear in seconds, which means any initial seed is easily obtained. So the game can be completely manipulated now. New TAS, Vykan?
Active player (255)
Joined: 12/13/2016
Posts: 352
Link to video First attempt at a WIP for H1 difficulty of this game. It's completed through Chapter 4 so far in just a little bit under 9000 frames, which puts me roughly 2 minutes ahead of Vykan's run (although I haven't forged the Rapier yet). With any luck, further optimizations will allow for a sub 10-minuted run!
Joined: 5/9/2005
Posts: 748
Always hoped you'd get into TASing. Felt it suit you better that relying on random RNG seeds.
GoddessMaria
She/Her
Reviewer, Experienced player (849)
Joined: 5/29/2009
Posts: 514
Location: Hell...
Whoa! That's quite amazing! I didn't expect this breakthrough but now that we have it, I am hyped for the next one!
Current projects: failing at life
Active player (255)
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)
Active player (255)
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.
Active player (255)
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)
Active player (255)
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.