Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I will add my voice among those who preferred the description being there.
Please bring the descriptions back.
Yes, that's the entire point of this post. I feel it is important enough that it merits posting by itself.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
mmbossman wrote:
From my standpoint as a judge, I think this a fine demonstration of a tough track. However, not knowing the game, is it only selectable as a stand-alone track, or can you play it as a normal track (i.e. included in a grand prix type game-play)?
To the extent of my recollection, it is stand-alone. It never shows up in a grand prix or largely anywhere else outside of the obstacle course selection. It is a unique track that doesn't loop like the races, and is not an arena for stunts or battles.
It's been a while since I last played this game, however. My memories have since faded. But remembering the trouble I once had just getting through is still just enough to make me enjoy this run.
As for the thought of a rejection based on the presence of a savestate (popular topic), that can be fixed by replacing the file with one that begins from power-on. This would take some effort by the author to try to have things sync up, however. If it can't sync, that would make things a bit more of a problem, then.
As long as you can match the time, I don't think anyone would notice the tire smokes are in a different pattern.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
StarShooter wrote:
Are you going to continue on that run next or are you gonna start over on a new?
New run. Something about Old/New PPU stuff. I want to use the Old PPU instead, which the abandoned run isn't in.
Going to be hard to remain entertaining throughout, however. Think you can dance like I did in the opening corridor?
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Here's my abandoned run, use New PPU. I want to redo some of this stuff in Old PPU, however. Just haven't gotten the will to do so, yet.
StarShooter wrote:
You can skip the transformation sequence after opening the seal to a corridor.
You do so by first press the select button (so you end up on the menu screen), then you press both select and start at the same time and hold that for a while (dont remember how many frames).
By doing so, you will save 264 frames, it normally takes 475 frames from that you transform to the next screen where the corridor starts.
Do the same thing when you have finished the boss on each corridor, at the time Miria (the hero) is transforming, do the the exact same thing as i mentioned and you´ll save 149 frames i belive.
I did not know this. Indeed, there are still things this game hides from me.
One question that pops into mind: Is there anything stopping us from keeping these buttons held while we're running through the game? This would avoid any further menu screen transitions.
StarShooter wrote:
I also found a trick which helps a lot on the blue optomon.
If you dont have any health left (the next hit will get you killed), and he´s about to drop another tentacle on you.
Usually, he wont drop it right away, he will follow you cross the screen and then drop it on you.
Most of the time, you wont be able to avoid it without stop firing and move away from him.
But, if you press select twice (first so you end up on the select screen, than again so the game starts) you can force the Optomon to drop the tentacle right away.
Now you can keep on doing damage to him without have to fly away from him.
This saves frames!
Is it possible to hide "inside" this boss? If there's enough room in there to keep your Saber in him while still avoiding the deadly attack (which is going upwards, now that you're in him), it might be a way to dodge a menu transition.
Another thought would be attempting to kill enemies in just such a way that you're near a certain score threshold and finally reach it by getting hit by one of those bullets when you need health. Nothing like a well-timed restoration.
(One time, when I was fighting the final boss, the battle dragged on, and my shields ran out. All hope was lost, there was no way I would survive long, and no item boxes in sight. Then suddenly, I hear the familiar jingle... One dramatic moment, I think.)
StarShooter wrote:
What is a 100% in The Guardian Legend?
I would say a 100% TAS of this game is all items, bosses and corridors.
My definition would be anything that affects the password, other than score or EE count. So this would include the items whose only purpose is to fully heal you, but are found in a few places scattered about outside of enemy drops. And these are marked off in the password as "collected".
And I agree entirely that 100% should also be a playaround -- As you say, what's the point of all that neat stuff we pick up if we never use them? A Saber-only run would get quite tiring, especially at a lengthy run such as 100%.
May as well visit a password room and show everyone what the password system is like. And how crazy-fast we can type stuff in there.
StarShooter wrote:
This is what my run looked like.
Going through corridors 1-10 (and ofcourse corridor 21).
Collecting only the items that was necessary, the two laser weapons (only the double laser maxed out) 6 red landers (total of 1200 powerchips) 5 defens shields, 17 in health bar.
I didn´t stop to fight any of the minibosses that gives you the attack item, i didn´t have the need for them and it would only waste precious time.
Damage chart is in this post. The Lv1 Saber deals 2 damage per hit with 1 ATTACK, and 4 damage with 2 ATTACK. The Lv2 Saber deals 4 with 1 ATTACK and 8 damage (only 5 versus bosses) with 2 ATTACK. In addition, we're also dealing 2 damage with our basic shots after the first ATTACK upgrade.
If you can remember anything about how long battles took, would having 2-damage or even 3-damage basic shots alongside 5-damage Saber hits be fast enough to make up for the time it takes to get it over just sticking with 1-damage basic shots and 4-damage Saber?
(Also, Lv3 Saber is no better than Lv2 Saber against bosses with 2 ATTACK, other than the ability to kill projectiles. For that matter, Lv1 Saber with 3 ATTACK is just as good as any higher level against bosses.)
StarShooter wrote:
It´s fun to hear that you are a bit interested in doing a TAS with me FatRatKnight, let me know later on when you have time and if you´re still up for it.
Either way, a TAS of this game must be submited here in the future no matter what, it doesn´t matter to me who record the movie, just as long as we all can work on it together in one way or another.
As i mentioned before, i´m still going to experiment on the game and see if i can come up with something new.
We'll see what comes when the contest is over, in around two weeks.
I'll try to think of something in my spare time.
Edit: Upload your run anyway, perhaps here if it works out. We'll let you know whether we can play it.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I wish to ask what the tricks are. I'm aware of a few, such as the fact that special weapons can only hit bosses once every 8 frames but still eat up any chance for regular bullets to hit on every frame. So make sure you hit the boss once every 8 frames with the special weapon (saber is easy to do this with) while shooting with regular bullets as well. Or the fact the same bullet is capable of hitting more than one target at once.
I have some interest in co-running this game. At the moment, however, I am a bit busy with the ongoing contest. When it is over, I'll have a bit more time for this.
In any case, I am quite familiar with the game, even going so far to crack the password through observation. I'd like to think I know every last detail in this game by now, but I wouldn't be surprised if there's still something I missed.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Currently, the contest has absorbed all my TASing time. Seems I didn't quite have as much spare time as I thought. However, I'm still pretty optimistic about the results.
When you find time, look over my plans, and see if there's any way to fit the new discovery into it. I've already listed three ways that we can solve the newly discovered flaw in my old plan, just a matter of comparing which one is the best choice.
Would switching Star to spot 4 save us any inconvenience at the cost of only one swap?
What's the cost of an extra random encounter for the sole purpose of + Agl?
Does starting MUTANT M in slot 3 and later giving that HAMMER to someone else work?
They all seem good, don't they? I'm not sure what to pick, but when this contest is over, I'll join in again to try out a few things once more.
And names. We're still undecided on them. 0-character names, or something so that the audience can actually identify who's doing what? Fastest possible says empty names, for obvious reasons. Do we go the fastest possible names? Now, I'm leaning yes, but I'll let you decide.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I just thought of one way to get instantaneous reaction from the user. But it involves input.get, seeing as joypad.get is one frame too late, no matter how you use it.
It also has a layer of inconvenience, since you need to set the controls from lua rather than use the emulator's internal set-up.
An example piece of code set only for player 1:
Language: lua
local Tbl= {
left= "G",
right= "J",
up= "Y",
down= "H",
A= "X",
B= "C",
start= "V",
select="B"
}
local Joys= {} -- Avoid reconstructing this table over and over...
--*****************************************************************************
local function Fn()
--*****************************************************************************
local keys= input.get()
for k,v in pairs(Tbl) do
Joys[k]= keys[v] or false --If it's nil, I'd rather take false
end
joypad.set(1,Joys)
end
Call this function alongside emu.frameadvance(). See if it is of any help. Still, involving input.get isn't a solution I would like to give.
So the earlier mess was just that we didn't communicate properly. I'm always learning ways to communicate, no matter how good I think I may be at this point.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
My statement still stands:
Language: lua
local function Fn()
local a= joypad.get(1)
if a.A then gui.text(1,1,"Yay!")
else gui.text(1,3,"No.") end
end
Use "boundary", it detects your button press one frame later.
Use emu.register before, it detects your button press immediately.
However, joypad.set works immediately in "boundary", and is late one frame in emu.register before. Either way you go, your problem still stands, naturally.
I have wrestled with how FCEUX deals with joypad, for several days, in fact, and I even wrote up details in here. I think I know my stuff. (After all this, I'd hate being called wrong without further investigation, which is why I'm defensive)
There is no solution to your problem for FCEUX as it currently stands. The implementation needs to change.
Regardless, while emu.registerbefore and emu.registerafter would get the user's current input through joypad.get, it is incapable of injecting new input in the immediate frame through joypad.set. Still a crippling problem.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Here's what's happening:
Joypad.get, in "boundary"*, will get the previous frame's input. While the user may be trying to mess with the joypad, it is not detected, instead preferring to read the joypad on the previous frame. And already, by the time the user joypad is read, it's blown away by the false you've injected.
It's still poorly implemented, making it impossible to read the user's input, transform it, and inject it all in the same frame.
I'm aware of a joypad.peek() function that has been implemented in a few other emulators. Especially useful for emulators that can run lua code while emulation is paused, as it tells you what the user is pressing rather than what was passed to the emulation. Would love it here, myself.
* My definition of "boundary" is lua code that precedes emu.frameadvance(). It is possible to place code that isn't in "boundary" by instead passing a function to emu.registerbefore, emu.registerafter, or gui.register. The code would be in their respective registers which I do not consider "boundary".
Regardless, while emu.registerbefore and emu.registerafter would get the user's current input through joypad.get, it is incapable of injecting new input in the immediate frame through joypad.set. Still a crippling problem.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Actually, I tried further tests, and I can't really make out any sensible logic in its inner workings.
Apparently, the only requirement for the first party member to go first is if the fourth member has a lower Agility. It doesn't seem to matter what sort of Agility the middle two have.
Also, setting Agility higher than BYAK-KO seems to guarantee going first. For some reason, I doubt this would happen in our TAS.
I'm going to stare at the World I RNG plans and make sure the first one that dies has a lower Agility than the one I want to win with.
So I see three possibilites:
- Kill the slow MUTANT M first, hence making him the fourth member after death.
- Reorder party so Star picks up GAZE instead, and don't give Agl+ to the first dead mutant.
- Fight an extra battle so Gaz gets fast enough to outspeed Fodd or Star, whoever dies first.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I have crashed into a problem. I am convinced that Star should take Gaz's place now, or else fight exactly one more battle to get the appropriate stats. Namely, Agility.
The game uses Agility to determine turn order. It's not the value itself, but the game sorts everyone in order of agility and picks randomly who goes next. Should the enemy have 90 agility, it doesn't matter whether our fighter has 8 or 80, it's treated exactly the same. Or so it seems.
I've been having problems killing the SABERCAT(s) before they kill someone. Who cares, I'll hack TELEPOR into Gaz to get out of jail free, and trudge on to BYAK-KO anyway. Alas, no matter what I do, BYAK-KO stabs Gaz dead before I can act.
A bit more hacking, and Gaz has some small bit of extra agility. Gaz goes first and creates art out of BYAK-KO. This "small bit of extra" happens to be exactly 1 point, from 7 to 8, and I go first.
I glanced at my dead allies' stats and noted that Gaz's Agility is right around where her fallen companions happen to hover around. Rather than hack Gaz's agility up, I decide to reduce the dead ones agility below that of Gaz.
Gaz went first. Because my dead allies lost some agility with my hacking.
For extra absurdity, I reduced Gaz's Agility to 3 and dead allies to 1. Gaz still went first.
I've got plans to rethink now... Either Star becomes the new Gaz, or the old Gaz fights an extra battle for agility. Any amount of agility would help.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Okay, I feel more awake than I was an hour ago. Time for useful replies.
Bobo the King wrote:
Sweet! Shared credit for a movie I contributed nothing to! I'll be sure to return the favor when I start my own run about a week from now. Even though I feel ready to run it now, I want to get a few obligations out of the way first so the run has my undivided attention. [...]
Shared credit stuff was a test to see if the names can fit. Again, they do. Besides, we're practically co-authors at this point. The amount of testing both of us went into making this run means both of us put a lot of effort into it, and both should take credit.
Bobo the King wrote:
[...] I'll try to make it complement your run in whatever ways possible (in particular, I'd like to see what extent of luck manipulation we can expect to achieve).
Personally, I'd rather see us go in a competition of who can manipulate the luck fastest at each moment where we need serious luck. The final movie gets the fastest roll between us.
Bobo the King wrote:
It's looking good so far, but I don't think I'll ever get used to those long resets.
And you shouldn't. Long resets = bad. We'll need to trim them down where possible.
Bobo the King wrote:
You said you'd explain things later, but I still feel compelled to ask about some slightly strange movement in the second half of the video. [...]
Laziness.
I may keep tight execution throughout earlier parts of the movie, but if all it amounts to is a test, I may end up just going realtime at the tail end and handing the resulting movie over.
Bobo the King wrote:
So can we manipulate enemy targeting but basically nothing else? If so, that's not so bad, since we might pick up ARMOR and make sure our invulnerable party member always takes the hit (assuming the monster/fiend doesn't get the first hit, in which case we're screwed).
We unavoidably pick up ARMOR on the way to TELEPOR. Besides that, we can also snatch up ESP while messing with POWER, which does take place before the character normally acts.
We also have another way to manipulate forced enemy encounters. At least, those that have varying numbers of enemies. ... Yeah, not a real good point.
Bobo the King wrote:
Lastly, how much of your time do you think the Dream Team Contest is going to take? [...]
A fair chunk. But I have enough spare time that I may be able to divert some attention here anyway.
This would be a fine time to determine the names of our characters for the real run (0 character names?), then manipulate the stuffing out of the first reset.
TheAxeMan wrote:
Have you considered equipping King Sword to help with Steward and his skeleton?
I haven't considered that. But now that I am, here's the tradeoffs:
The equipment screen and ability screen are separate submenus. It will eat up several more frames like that. Also, when it finally comes time to insert sword into statue, sword is in inconvenient place, requiring me to exit the item submenu to get the sword in another submenu.
Regardless, it's worth checking to see if changing equipment beats the animation of two RAPIERs.
Now I'm taking a close look at my RNG plans...
First reset, we need these:
C30B:51 (GAZE)
C36F:5F (Ability in slot 3)
60 outcomes. Not real hard.
Here's what we'll encounter:
C343:58 (FG; 8 outcomes; Open land first gets us GOBLIN. Good. Must follow with forest encounter for another GOBLIN)
C343:59 (GE; 52 outcomes; Open land first gets us LIZARD. Bad. Forest gives GOBLIN then ALBATROS.)
If we walk exactly 6 steps, we can pick from either one. But I like the ALBATROS encounter because Star can one-shot the bird with SABER. Besides, we generally need less luck-related effort to get these encounters.
Recall, the purpose of these battles is to give Gaz GAZE and then Mana+5.
Second reset is for TELEPOR. Details:
C343:F2 (GGF. All others that can give TELEPOR in 4 battles nails us with an F encounter in the forest, giving ZOMBIE)
64 possibilites.
Well, that's easy. We have 64 possibilities. Except the details are a bit messy...
C30B:EB C36F:F9 (Slot 1 TELEPOR to second member. 8 possibilities)
C30B:EA C36F:F9 (Slot 1 TELEPOR to third member. 4 possibilities)
C30B:EA C36F:F8 (Slot 3 TELEPOR to third member. 52 possibilities)
I particularly like the elegance of the first option, as it means no form of party switching is needed. Pow isn't expendable until after BYAK-KO this way, but his fighting usefulness ends two battles earlier, after Monster Trio. TELEPOR is slightly better on Gaz for this reason, but it doesn't seem critical at a glance. Keeping with the first option really does seem best, but I need to get on those tests.
However, it's a pretty narrow range. The manipulation required will be tough. Thankfully, the range of spare steps is pretty generous -- feel free to save a few steps early if TELEPOR is that painful.
Then POWER:
C30B:95 (POWER, BARRIER, nil, HP+)
C30B:96 (Def+3, POWER, BARRIER, nil)
C30B:97 (nil, Def+2, ESP, POWER)
Second and third possibilities are good for POWER, as it means no switching again. It also means the third character, Fodd in this case, eats up both BARRIER and ESP.
POWER is stuck in slot 2, no matter which of the three your pick.
BARRIER can be in 1 or 2 if you pick the first possibility, but only 2 if you take the second.
ESP can be put in slot 1 or 2.
I'll hide the major analysis here, but there are 52 possibilities for a slot 1 ESP with E encounter, giving ALBATROS in forest goodness. There are 8 for the same ALBATROS encounter with C30B:96.
Battle of KINGSWRD
I have chosen C30B:14 here. This gets us Mana+4 while we beat the stuffing out of stare at KINGSWRD until he drops dead. Unfortunately, this leaves a chance where running away fails. C323:18 is what we'll need, and that leaves 24 possibilities. Not as rough as TELEPOR to second member by any means, but still not easy.
I haven't searched for equivalent values that still let me run from the first battle to give KINGSWRD a good look. If we don't run, KINGSWRD smacks someone around first, hopefully Star, but while it is true that I had a little cash to spare for the inn and BATTLE sword, it's only 33 GP in my test. 34+ damage forces me to sell something to make up for the loss in the future inn trip.
Then it's time to slay STEWARD:
C353:94 and C30B:89 shows one way to get a solo SKELETON with a Mana+5 thrown in. Problems: The Mana+5 is given to the first mutant, requiring a switch, and there's only 8 possibilities to even get it.
C353:FC and C30B:F1 is another thing we can do. Also gives solo SKELETON with a Mana+5 given to the fourth mutant, right where Gaz is. Again, only 8 possibilites. If the luck is still too horrendous, we have another 120 possibilities around here with C30B:F2 or C30B:F3 and C353:FD, but this requires reordering the party.
So, effectively, we have a wide 128 possibilities that require moving the party around, and 8 that has everything fall neatly in place. Take the 8 if possible within the timespan it takes to switch party members, but the 128 seems to give a nice cushion in case the RNG hates us.
Either way we go, however, the number of steps to next encounter is few. Very few. We almost certainly want to reset after the battle against SKELETON and manipulate a Mana+5 in whatever slot Gaz happens to be in, and a better "steps to next encounter" RNG.
Still not sure how KING sword does, here. We might want to TAS the two ways to see which one's faster, but this has a problem that we need to manipulate luck in between equipping it and actually trying it out.
These are the ideas and details I have from the RNG test through World I. ... Soon, someday, I'll get a set of tests up for the required battles stuff later.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I started running realtime right around King Shield. I left virtually all mistakes in there, as I don't intend to keep this run anyway. Any strange pauses, mysteriously misplaced menu stuff, or extra walking can be filed under "didn't bother being TAS-precise". This is not a real TAS run, simply another test. ... As for why I chose to be TAS-like earlier in the movie... Yeah, good question. My habits, I guess, but said habits did show me how to use TELEPOR two frames faster.
I simply wanted to try out an RNG path with TAS-like precision minus the TAS-like speed in luck manipulation. At some point, once we finalize the RNG route, we'll just avoid those long thousand frame pauses I do. Besides, one more test gives me an excuse to give descriptive names to our characters so we can identify who's doing what.
Again, I'm simply trying out this RNG route. And it seems to be a pretty good one. I just need to walk through the remaining battles with non-existent stats and narrow scrapes to see if they're still all possible to beat.
As for the author field, I hope the description that comes right along with the movie tells you that I'm seeing if our full identifications fit. They do.
The GAZE manipulation looks like the way to start. Just find some way to trim off the long pause, and then just fight the two battles, first for the GAZE, the second for the Mana+5. The alternate enemy group we could fight is a ZOMBIE in the forest to start. ... I'm thinking what I got in my run is the best choice...
I was worrying about the HP gains for a bit (means I pay more at the only inn I stay at), but at the end of the movie, I still have enough money for a BATTLE sword.
Between these two, I picked the second one. This allows me to put TELEPOR into Pow without any party switching. In addition, Gaz picks up Mana+2, but that should have happened either way I go. ARMOR is pretty unavoidable here anyway. Star is probably the best one to get ARMOR.
Star> --
Pow > Def.+2
Fodd> 1]ESP
Gaz > 2]POWER
Star> Def.+3
Pow > 2]POWER
Fodd> 2]BARRIER
Gaz > --
When giving my mutants POWER, I looked for any way to get a one-battle POWER that didn't include putting it in slot 2. This was a failure. Regardless, there's still a way to equip my long-term mutants with POWER still without making any switches. These are the two I picked, though there is a minor variant in what slot ESP or BARRIER can be put in, but it's late here and I don't feel like checking too deeply.
I start with the first one, as it allows me to fight an ALBATROS while still milling about in the forest. Star can one-shot the ALBATROS with the SABER, so I took that.
Based on when the first and second encounters would take place, I decided POWER would be best obtained from the random encounters. You fight one early, then get a long streak of no battles following. I feel this is as good a reason as any to obtain POWER from a random encounter instead of swapping a Mana+5 we'll get from later required fights.
To the KINGSWRD, apparently, by running away from one battle, I can get Gaz to go first. There is one particular RNG value I found that lets me crash into one battle with a high step count after that, while still handing over Mana+4 to Gaz. The Mana+2 earlier boosted Gaz up just enough that a +4 boost later won't force me to fight an extra battle for the needed BYAK-KO killing 46.
As for King Shield, that's where I stopped bothering to TAS with any accuracy and started realtime running, forget fixing any mistakes. I recall finding only two RNG values that gives Mana+5 and a one SKELETON battle. Beyond that, I probably should have simply added a reset on the way to STEWARD to get a Mana+5 inline for Gaz rather than switching the party around. There's a lot of no-encounter steps between the two, meaning save at any other point in between if saving at one location didn't give a speedy RNG manipulation.
Overall, I think this might be the sort of RNG path to go for. Shorter manipulation pauses, further analysis when I'm less tired, and cleaning up the disaster known as "not a TAS" around King Shield castle. The pauses in that battle have no meaning whatsoever -- I got lazy. The same damage would be done if I had attacked as quick as possible. Victory would still take place after those three hits just fine.
But as few battles that I get into here, it's all for nothing if it means I can't win every battle later on. No need to bother with keeping optimal if all I want is a set of characters to beat later foes with. Why did I bother early in, anyway?
SEI-RYU is beatable with the party I ended with, and miraculously enough, ESP is proving its worth in defense for Fodd. She can still die at round 2, lived up to the fodder I'm naming her after. ARMOR is also a viable defense, but only if Star sits in the first spot in order to shoot it off early enough.
The MOSQUITO battle is more trouble than I expected, but oddly enough, manipulating three of them lets me get through them without a scratch, while having two or only one results in a death that I'd rather not have happen here.
I've still got numerous other battles to check, though. I'll record each of them as I climb through the tower, don't worry. I'm using a script to get rid of random encounters to make the trip easier.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
World I test.
I maximized the amount of steps between battles. Up until the very end when I said "It isn't worth optimizing crazy-lots."
The good old trademark "Very Long Resets" like in my test run can be seen here. Except extra long in the last case. I don't care -- I'm looking for the right RNG, just to see if it can be done at all.
Turn order happened differently than what I'd like, as I'm not giving GAZE to Star.
KINGSWRD battle had me run from a previous battle first.
Solo SKELETON was more durable than I initially thought.
I didn't bother at STEWARD. Other than making sure I didn't overwrite POWER with P-BLAST, of all things. Fodd holds P-BLAST at the end, by the way.
I'll copy this party into various parts of the 40-minute test run to see if this can work out. Maybe, maybe not, that's the point of a test. No matter the optimization, if it won't work, it's all a waste.
I'll explain stuff later when I feel like it.
Edit: Oh, battles can be manipulated. One aspect of it, anyway.
Enemy targeting: Address C35A
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Something like "Company: Capcom"? Handing out a category based on who makes the game is another way to divide them, and might encompass useful sets. Would be good to keep some ideas around just in case.
But first, it might be a better idea to focus on a list that doesn't shove several kilobytes per game at the local user who picks some 490 item list. Hello there, megabytes of download just to browse the list. This is the main purpose why I created the topic, but different ideas of having it happen are welcome... To the site's staff, I hope.
I have an aversion to using the lists for the reason of what sort of load is to be expected. Which defeats the purpose of having them (for me).
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
sgrunt wrote:
So, something like [wiki ListAllMovies]this[/wiki], but sorted first by platform?
More like some sort of filtering system than a sorted list of all movies. Besides, suppose I want to see the platformers. Those tend to be on multiple systems, and it's not exactly easy to find all of them if sorted by system.
Warp wrote:
1) Create an alternative smaller version of the current movie info box, containing the same relevant info but in a more concise way (ie. taking much less space), possibly also with a smaller screenshot. This way long lists of runs would take less space and would be nicer to browse.
It is one idea. Smaller pictures is certainly easier on the bandwidth, and shorter descriptions, perhaps no more than three lines, will make the list easier to browse through.
Even better if such a list can have each game with an arbitrary number of published movies all fit in exactly one line. (With a link that points to all relevant movies as seen in my example list, particularly the Adventures of Lolo 2 example. Something like that, anyway)
Warp wrote:
2) More movie categories [...]
More filters are fun. Implementing new categories will only get to be an increasingly longer task, as it appears to me, but once in place, gives a new way to filter things.
It would also be nice if there is a way to apply multiple filters. All platformers on the NES, for example, but this might mean something less static than "here's a list of games under this singular category".
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
There really needs to be an easy index. The search engine is great, but it tends to be that you generally know what you're looking for if you're going to use that. Suppose I can't think of any particular name to look for...
So, for those who just want to browse around, opening up a list of movies for a system or category brings up... A mess. Loading every single image of every non-obsoleted movie for a system, or worse, a rather bloated category (Genre: Platform (490)) in a single page is not something that appeals to me. Not to mention the site load this implies.
There's also the fact there's a lot of vertical space being taken up in a list. About three or four game names can be displayed on my screen at once. Looking through it isn't exactly speedy.
A list of simple links is preferable, to me. If it can be automated, all the better. I believe maintaining such a list manually is not a trivial endeavor. And should someone not like how the list was manually formatted... Anyone up to going through the entire list just to realign things?
NES
* 8 Eyes
* A Boy and His BlobALL
* Action 52
* Addams Family, TheALL
* Adventures of Bayou BillyALL
* Adventures of LoloALL
* Adventures of Lolo 2ALL
* Adventures of Lolo 3
* ...
Something like this.
Note the game with multiple unobsoleted publications has the main link pointing to them. Adventures of Lolo 2 being the example in my short list.
If there are any reasons not to have such a list, such as difficulty of automated implementation, I wish to hear of it. If nothing else other than to understand why such a list isn't available.
If there is already such a list I'm asking about, point me there. I obviously can't find things myself.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Hmm... Good job searching. You have pretty good methods.
It also saves me time from having to check myself. Part of what being in a team is for: Several different tasks can be done at the same time with multiple people working at them. We can also "compete" at a part using our own methods of optimization and produce something that an individual player wouldn't spot.
Besides that, there's also the fact we're suddenly working on this furiously, instead of what happened when we sort of dropped out.
I was hoping for a 2-mutant possibility without having to shuffle members around. That way, one can kill off a spare mutant at the EVIL EYE battle. Seeing as that's not the case, this generally calls for some more testing to see if any can work out. I'm hoping to get by with the original four mutant party with no replacements except for the MUTANT M / HUMAN M from the northeast town, which doesn't leave me with anyone who can survive a single attack, except for said replacement.
I think it's time I started on another test run of sorts. First, I need to check the RNG path in World I. Then it's off to SEI-RYU to double check that battle. Then, naturally, World III battles. Then I will finally check the difficult World IV stuff.
I base my values on what they roll to, rather than what they were prior to the roll. Basically, our order of operations appear to be different:
Me:
Increment(RNG)
Use(RNG)
You: (?)
Use(RNG)
Increment(RNG)
As such, if the "Do you encounter something next step" RNG happens to be 0xFE before you step, then after stepping, you say the game uses the 0xFE number (as that is what it was) while I would say the game uses the 0xFF number (as that is what it will be) to trigger the encounter. The way we have our charts reflects this.
My habits came from running E.V.O. the Search for Eden and, more recently, Etrian Odyssey, as the games use the updated RNG I see in memory in the following frame rather than what the number was prior to rolling. (These actually roll numbers visible with memory watch rather than incremental +1 like we see here)
Who is "right" in this case? No way to know until we find a dead simple algorithm that prefers one or the other when generating it. There's a good chance the game does use a simple algorithm from the smallest base, which can only be more complicated with a "shifted" chart.
Either way we look at it, just shift the chart, and it's correct suddenly. That's obviously no proof who's right.
As for starting numbers, we both can see that on the next reset, this RNG address hits that number, and we both would call it the same number. The moment the game needs the RNG is where we differ.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Considering that there's a push to update VBA in time for the Dream Team Contest, I would like to avoid actually making a serious start until it's updated.
Still, I'll be making a low-effort test of a route I posted several hours ago. At least make sure I didn't become insane with all that planning.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
I've been sent a message as well. I'm not even a Super Mario World TASer (although I did give a decent suggestion for someone in the topic, and I do follow a few things).
This is what I see currently happening in Super Mario World TASes:
There is the 100% version being worked on, an improvement over what we currently have. With new glitches being found, there is potential for Fire Flower being used a few times for these glitches, should they save more time than staying with the cape.
In addition, I'm aware of someone working on a playaround. From a piece of the run that we got to see, there was some use of Fire Flower in an earlier stage than other runs, though primarily used for glitching, as is in the nature of the run. As this run is not constrained to time, I wouldn't be surprised if the Fire Flower gets decent screen time.
Whether they will be seen more often here than in other runs, that remains to be seen. I am not aware of anyone with particular interest in Fire Flower only.
Finally, a point that I am not fond of...
Asking once for something to happen is fine. If you have to ask twice, consider trying it yourself for a bit. TAS projects tend to take a very long time, and even if someone does agree, it's not often something that will be finished overnight. Even if we are patient, we tend to have other things to do. We may be a TASing community, but it's not like the only thing most of us do is TAS.
You've asked several times, and even sent messages to people (such as myself). I don't have anything to offer other than what I said in this post already, and anyone viewing a topic like this that did have interest probably would have chirped in by now. This place doesn't seem to have people with much of a desire to fulfill your request, and I do not know where to look for such people.
DarkKobold gave you a final warning. Please take it and quiet down. Asking us again won't speed anything up other than your own loss of access to this site.
(I don't mean to act like a site staff, as I am not. I just wanted to say these points in my own words.)
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
When I say detailed, I mean detailed...
World I detailed plans:
- Party of four mutants:
MUTANT F (Star)
MUTANT M (Pow)
MUTANT F (Fodd)
MUTANT F (Gaz)
The SABER has an attack of 6, where the LONG sword has an attack of 3. Also probably gives us a better turn order against the random battles. I doubt I'll use this mutant beyond World I. For purposes of giving recognizable names while explaining my route, I name her Star, our starter.
The HAMMER is a Strength-based weapon, which a MUTANT M starts with. POWER exists and sets our Strength to 99. There's no equivalent for Agility. He shall sit in slot 2, given the name Pow, for POWER abuse.
The MUTANT F in the third slot will follow Star in the turn order in the second random battle. M will not. Seeing as I plan to crash into the occasional second random battle without an intervening reset, striking the enemy again before taking a hit sounds wonderful. Other than this purpose, however, this mutant serves no real purpose other than to die a horrible heroic death later. As such, her name is Fodd, our neighborhood fodder!
Finally, the MUTANT F in the last slot will follow Star in the first random battle following a reset. M will not. With the lack of significant damage dealing options, enemies will need to die from two hits, which SABER + RAPIER attacks can do. As GAZE shows up as the fourth bonus in one RNG instance, this mutant will pick it up after the first battle, giving her long-term use, so we should pack POWER in with her. Her name in my route planning here will be Gaz, the GAZE user. Don't worry about the starting 5 Mana instead of 6, this point is moot later.
I don't expect to actually name them such in the actual run. I just want to have recognizable names in the text here, to speed up communication here rather than naming suggestions.
- Manipulate for 0x51 GAZE
--
df+1
hp+14 / 11
AB 3 GAZE
st+3
Shuffle
hp+13 / 14
Mana+5C36F must be 0x5F!! (0x60 puts it in slot 2, where POWER is going)
Do we reorder the party so Star picks this up? I'm thinking not, the random mutant with 5 Mana instead of Star's 6 is no problem, looking at the RNG path for TELEPOR.
Fight one more battle before resetting. There's a handy Mana+5 in the perfect spot where our GAZE mutant needs it anyway.
It may be prudent to reset after taking at least 6 steps out into the overworld. Unless you think you can manipulate C343 as 0x58 by resetting at the title screen, which is not an easy piece of manipulation to do here.
- Manipulate for TELEPOR
df+2 - - - - - - - - 0xEA
AB 3 ARMOR - - - 0xEB
st+5
Mana+2
--
--
hp+12
st+1
--
ShuffleMana+1
--
Mana+2
df+2
AB 1 TELEPOR
df+3
AB 4 STEALTH
C343 must be 0xF2!! (We'll have no way to deal with a ZOMBIE in either of the two forest encounters otherwise)
The first two battles will be in the forests, GOBLIN and GOBLIN. The third will likely be in the cave, against ALBATROS. The fourth will be from the P-FROG bandit. That's where we can finally get TELEPOR.
We will have two long-term mutants to hand TELEPOR to. One possibility is to hand it to Gaz, but this requires putting TELEPOR into slot 1, meaning C36F must be 0xF9, making manipulation tough. Regardless, Gaz needs at least one of the various Mana ups found in this path.
On the other hand, if we find that we can't manipulate a slot 1 TELEPOR effectively, allowing for the possibility of the slot 3 TELEPOR with C36F at 0xF8 might make things slightly easier. Pow will need to take this one, as otherwise it will overwrite the slot 3 GAZE of Gaz. More switching will need to take place, as we'll need to give the Mana+2 to Gaz.
And that ARMOR, that's going to burn menu time if Pow ends up getting it. Ouch. Star is a much better candidate for it. The most optimal RNG here seems like C30B = 0xEB and C343 = 0xF2, but this is mighty difficult to get a hold of.EDIT:Actually, now that I think about it, it might be better on Pow anyway for the fact Star needs to use her SABER more often than Pow needs access to his HAMMER. Hard to say what the best RNG is, really.
P-FROG eats a GAZE. 4 left.
- King Armor
Use freshly gained TELEPOR. Yum, fresh TELEPOR.
Aim for POWER overwhelming from the random encounters, so Gaz and Pow aren't weak and worthless at SEI-RYU. I think there are three possible values for C30B that allows a one-battle POWER, but as "luck" would have it, I can't find any of them that isn't placed in slot 2.
Seeing as we already fought 5, that's a total of 7 random encounters for 280 G. That should be enough, but cutting it fairly close for the BATTLE sword later.
- King Sword
TELEPOR again. 3 left.
He eats a GAZE. 3 left.
Unlike what I was previously thinking, no one dies here. I instead now plan for Fodd or Star to die at SEI-RYU, and the remainder to die at Monster Trio.
- King Shield
TELEPOR a third time. 2 left.
C353 must be one of these: 1B 2A 2B 2C 35 4E 52 64 6D 7A 8D 94 97 99 9F A1 A2 A3 A6 A7 AF C4 C5 C8 CA E0 FC FD FE
This is so we face 1 SKELETON. Not two, for that is one too many. And certainly not with an ASIGARU, for that, too, counts as an enemy, and thus there are two, and still one too many. A SABER and a RAPIER together can take on one, but not two.
STEWARD can just die to the usual SABER + RAPIER, I think. Maybe SABER alone, considering that wonderful 1 defense.
I'm hoping Gaz can pick up at 9 or 10 Mana from these two battles without an intervening reset or party reorder. But if there's no such RNG, an intervening reset seems to be the way to go.
- To the Tower
TELEPOR once more. 1 left, but it's restored by local inn.
Don't forget that NPC. We will pay dearly later on if we forget to talk now.
At the tower door, my test run uses a sphere then immediately saves and resets. I found this to skip the message and flashy effect, but unless we're manipulating encounters away at the same time, this won't save much time.
Seeing as we have 3 GAZEs to spare before we hit the inn, we can and probably should use 'em on random encounters... Assuming Gaz can outspeed Star when using GAZE. On the other hand, if the GAZE animation is long enough to exceed the length of the SABER and RAPIER animations together, then forget this idea.
Finally, here's a small piece of code I made for myself to scroll through possible RNG values:
Language: lua
local RNGstr= "???.??"
local RnSet= 0
--*****************************************************************************
local function AdjRn(Val)
--*****************************************************************************
RnSet= (RnSet+Val)%16384
local v= RnSet/64
for i= 0, 0x7F do
memory.writebyte(0xC300 + i, math.floor(v))
v= v+(9/64)
end
RNGstr= (string.format("%3d",math.floor(RnSet/64)) .. "."
.. string.format("%2d",RnSet%64)
)
end
keys, lastkeys= {}, {}
function UpdateKeys() lastkeys= keys; keys= input.get() end
function Press(k) return keys[k] and not lastkeys[k] end
--*****************************************************************************
local function UserAdjustRNG()
--*****************************************************************************
UpdateKeys()
if Press("numpad7") then AdjRn( 0x0001) end
if Press("numpad8") then AdjRn( 0x0040) end
if Press("numpad9") then AdjRn( 0x0400) end
if Press("numpad1") then AdjRn(-0x0001) end
if Press("numpad2") then AdjRn(-0x0040) end
if Press("numpad3") then AdjRn(-0x0400) end
end
while true do
UserAdjustRNG()
gui.text(1,1,RNGstr)
emu.frameadvance()
end
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Bobo the King wrote:
[...] However, getting the shoe will require us to go to the Equip submenu, which is technically out of the way. I'm comfortable waiting and seeing.
Actually, it's scarcely out of the way at all -- We need to equip the SAW on him anyway, which we should be able to do in the same time.
Bobo the King wrote:
As for who should equip the shoe, you seem to have indicated a Human would still be marginally faster. My only thought to the contrary is that we might use one of the overworld SU-ZAKU battles to pump a Mutant M. Luck manipulating out of them will be a pain anyway and just getting into a battle with him will be time consuming, even if we run. What's not so great about this plan is that even overworld SU-ZAKU has that fancy fadeout that costs a second or two. I doubt it'd be worth it, but I say it's worth consideration.
In order to slay overworld SU-ZAKU, we need to use the SAW. To use the SAW in 1-turn fashion, we need 75 Strength. The fact we're rest assured not at 75 strength (what's the point otherwise?) and the fact this might only save one turn for a MUTANT M at the real SU-ZAKU battle makes things sort of skewed towards "not worth it".
In any case, I haven't actually checked the mutant bonuses in great detail. For EVIL EYE and MACHINE, we need two strength bonuses that total +6 or higher in the span of four "steps" while still having enough walking/biking steps to avoid encounters between EVIL EYE and MACHINE without a reset. Even if it is possible, a MUTANT M is still stuck throwing two turns at the real SU-ZAKU. The trade-off comes from the fact a HUMAN M is instead going to a STRONG shop, but will kill SU-ZAKU in a single turn.
... I sense a split in the path that only converges when we finish World IV, where we'll need to test two paths. Unless it is discovered we can't get at least +6 strength in two battles (with a two-mutant party) without a reset.
Bobo the King wrote:
Up to this point, we've implicitly assumed that we need to buy STRONG potions after we get the Human M. [...]
I'll make it explicit -- HUMAN M has 54 Strength. We need 75 to kill in the first shot using a SAW. Only way for HUMAN M to build it is to buy STRONG to permanently get the missing strength, or get equipment with bonus strength attached. STRONG has no effect on anyone who isn't HUMAN M or HUMAN F, so we can't build up a MUTANT M using STRONG.
There. Now anyone who doesn't know why we're bothering will now know.
Bobo the King wrote:
[...] I'm pretty sure they can sit in our inventory through most of the game without any trouble.
6 STRONGs? Out of our 8 inventory slots? If you're going to suggest we equip some of them for added inventory space, that's even more menu time!
Regardless, here's a list of items I expect to have throughout the game, in rough order we pick them up:
- SABER (Comes equipped with starter MUTANT F)
- HAMMER (Comes with MUTANT M)
- RAPIER/HAMMER
- RAPIER/HAMMER
- KING armor
- KING sword (Thankfully, we'll lose these things quick)
- KING shield
- BATTLE sword (For the quiz. Might replace a weapon)
- AIRSEED
- BLUEKEY
- RED ORB
- BLUE ORB (orbs consumed on route through tower)
- WHITKEY
- STONE book (Might replace a weapon)
- SAW
- ROM
- GETA shoe
- BOARD
- BAND helmet (After the northeast town stuff)
- ERASE99 (Also late)
Granted, we can throw away a piece of junk at a tiny cost to frames while using the spheres, but that only takes care of two of the items. If we include everything out to ERASE99, we'll have collected 9 items. Include the 6 STRONGs, and you'll find that 15 items don't fit in our main inventory, only 8 does. Okay, we discard two of them in the process of using our spheres, that leaves 13.
Heck, even if we have just the SAW, GETA shoe, and ROM -- Oh snap, 3 items. We broke the 8 item limit just there, if we include 6 STRONGs. Losing the SAW means we lose our only means of slaying endgame bosses, the GETA shoe replaces one shopping trip with another, and ROM is a mandatory item. We can't equip the SAW and GETA shoe on the HUMAN M, as said human isn't in our group yet when we pick up the GETA shoe.
As you see, an inventory limit of 8 items really gets in our way here. How should we best handle this? Discarding old key items will happen, and two of the discards are "free", however.
I wouldn't rule out the BAND helmet or ERASE99 as inventory-related trouble -- It's possible we decide to use STRONG after the MACHINE battle. Why not right away? This would mean we avoid equipping the GETA shoe, or else our strength exceeds the MACHINE's defense, and then no more SAW. We can still use 2 STRONG, though, for 4 free slots. We can also throw away the ROM and BOARD before we pick up ERASE99, but that also requires a special trip into the inventory screen.
... And all MUTANT M sacrifices is a round at the real SU-ZAKU... If the +6 strength is possible from the EVIL EYE + MACHINE no-reset path. The HUMAN M complicates things!
Bobo the King wrote:
[...] it seems we have no reason to shop anywhere except Northeast Town. [...]
Well, that does simplify things in terms of where to shop. HUMAN M is still pretty complex, however.
Bobo the King wrote:
•STEALTH: Apparently offers a greater chance of striking first in battle. [...]
As far as I know, the "greater chance" is the difference between having it happen and a zero percent chance. Yes, without STEALTH, you can't strike first. I may be wrong here, though.
Also, starting as a mutant gives us this ability right from the start.
Bobo the King wrote:
•BARRIER: Adds a paltry ten defense to all characters, but since our defense is so low to begin with, that may make a lot of difference. I don't see it helping unless we find ourselves in a rather unusual situation.
Said "unusual situation" would be that we are caught in a battle that can't be done in one turn, we lack any decent (read: still living) humanoid shields, and that 10 defense lets us survive with our paltry HP. I fear this situation won't ever come up.
Bobo the King wrote:
•QUAKE: You're already aware of and is our preferred "alternate" spell for dealing damage. It may be our primary spell now that we're moving away from P-BLAST.
Although, I'm moving away from *all* direct-damage spells. Roughly 26 Mana for SEI-RYU and 36 for Monster Trio won't really cut it.
Bobo the King wrote:
Honestly, I don't think we'll be using any of these, but we should probably know about them anyway because we may need to fight an extra battle at some point. If we can squeeze a useful ability out of it, all the better.
I'm pretty sure we won't need any of these. But it's still good to know our options.
Bobo the King wrote:
When do we manipulate luck and when do we run?
An optimal reset takes a shorter time than an optimal run away. Hence, it is almost always worth a reset (and manipulating the "steps to next encounter" RNG right) over attempting to run away.
That's my answer to this question.
Bobo the King wrote:
Just one: Will you have enough money to buy the Battle Sword? [...]
I assume I will. This will take further analysis, though. I might need to substitute a World I encounter with an early World II encounter if money proves to be a problem.
Bobo the King wrote:
Sounds like you know pretty well what you're doing. For the moment, I'm going to work on decoding more RNG addresses and figure out what's going on with ability shuffling. After that, I think I'll be ready to join you in running this game. That is, unless you have other ideas to test.
Currently trying to think through a World I plan that minimizes switching party members around, and maximizes the amount of steps we can adjust at the RNG for a particularly difficult TELEPOR manipulation.
I'll post the full details later. This post is getting a bit bloated already. Since you are studying the RNG, peek into that for strength bonuses, alright? Assume a two-mutant or three-mutant party, we need at least +6 in two battles without a reset, preferably without switching party order in between. Even better if the bonuses show up in the last slot. In addition, report step-count until next battle.
Encounter rate of EVIL EYE floor is 5 (0x2D 0x7C 0x96 0xA1 0xFF), while the encounter rate of the floor above that is 4 (0x7C 0x96 0xA1 0xFF), in case you need this information. 17 steps between EVIL EYE and the next floor, and 56 total steps from EVIL EYE to MACHINE, by the way.
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
pcsxrr... Another one I can't say I'm familiar with.
The button2:draw(); line is optional. My script already handles drawing stuff. A line in my file says Don't bother using it, it's automatically called.
But that doesn't solve your problem... Give me some time to think of something...
But to start with, uncomment the print(Z) line. I'm running your code pretty well from DeSmuME here. (Although I may want to double-check my recolor routine)
You might want to throw in gui.text(1,1,keys.xmouse) in _FRK_Btn.HandleBtns(). Should at least show the X position of the mouse. This will take further thought...
Editor, Experienced Forum User, Published Author, Skilled player
(1174)
Joined: 9/27/2008
Posts: 1085
Emulator?
Does the button get highlighted at all when you mouse over it? That is to say, does it have any visible difference in color when you're moving the mouse on and off the button?
Was the emulator paused at the time?
I recall there being some sort of emu.wait and gui.redraw somewhere. I don't have the lua documentation handy for Gens, and I'm completely unfamiliar with the emulator. I recall there being something like gui.redraw and emu.wait functions for Gens, but it would take several lines to have them implemented. I haven't seen similar functions for Snes9x or VBA, however.
In particular, the example script I have for Super Mario World will not react to the mouse if you have the emulator paused. Its magical transformation can only happen when the emulator is running, thanks to the fact gui.register is never called more than once per frame, and Snes9x 1.43 and 1.51 is missing the critical components to call lua code while paused.
Again, I'm asking for what emulator you're using. I'm going to guess Gens.
Edit: This is a stab into the dark, but if you're using Gens, and my wild guessing is any good, this piece of code might actually do something:
Language: lua
local lk= input.get()
--*****************************************************************************
local function BoundaryKeyScan()
--*****************************************************************************
local k= input.get()
if k.A and not lk.A then --Pseudo frameadvance
NoAdvance= false
LuaPause= true
end
if k.B and not lk.B then --Pseudo pause/unpause
NoAdvance= false
LuaPause= not LuaPause
end
lk= k
end
while true do
NoAdvance= true
while NoAdvance do
gui.redraw()
gens.wait()
BoundaryKeyScan()
end
repeat
BoundaryKeyScan()
gens.frameadvance()
until LuaPause
end
Gens? Stick this at the bottom of your code, getting rid of any while true do gens.frameadvance() end you might already have in there.
I found the documentation. I just haven't gone far enough to actually find a ROM someplace and run the emulator yet.
All this, currently, is a wild guess. First, it hinges on the fact you're using Gens. Second, it also hinges on the functions doing what I think they do. Lastly, the fact I can't even test this stuff myself at this hour means there's no way for me to find out on my own if what I typed even works.
But here's something to try anyway. The chances it works are low, but I'll be sure to reward myself with a pat if it works out anyway.