Where is it first impossible to beat the game without the B button? Is there a jump in a particular level or something that cannot be cleared?
Also, if a warped run is possible using only B to open the inventory, then it might be a better idea to do a warped B-less run on SMA4 (GBA), where L opens the inventory. This would just be to make the run definition absolutely clear as "never presses B at all" instead of "only presses B outside of levels" which is a tad arbitrary.
That's actually really helpful; I did not know that stat-lowering moves like Growl could still fail in Gen II, nor did I know that only the opponent's stat-lowering moves could miss.
On a wild guess, when people stop playing Gen V on Wi-Fi, so that abusing the in-battle RNG could no longer be really used to cheat anyone in Wi-Fi battles. Which will probably be in like 2 more generations or something.
Ok, so say I want to get 5 critical hits in a row.
In RBY/GSC/RSE/FRLG/DP, I could potentially make a savestate right before each turn, and reload each turn individually to try to get a critical hit. Given a 1/16 critical chance (I know RBY's are different), this would take around 80 reloads on average.
In Plat/HGSS/BW/B2W2 and presumably all future games, the only way to get 5 critical hits in a row is to manipulate the starting seed before the battle begins. This would on average require searching through 1,048,576 (16^5) seeds.
Is the above accurate? If not, what would be a more accurate description?
I thought the Battle RNG from Platinum onward only advanced when it was called? How do you call it without using a move?
Is it possible to manipulate and get, say, all critical hits in Platinum or HG/SS?
Now that I think about it, you're probably right. I've never experienced any high "But it failed!" rates on stat-lowering moves outside of the first battle.
EDIT: I hacked a wild Pokemon to know only Tail Whip, while I spammed Harden so my own Defense would never reach -6. Tail Whip never failed through all 30 of Harden's PP.
I hacked the rival's starter to know only Tail Whip, while my starter knew Harden and used it 30 times. Tail Whip failed around five times.
Confirmed?
I just tried it myself, no tools, on a Red ROM. I restarted the first Rival battle over and over, keeping track of each time the opponent used a stat-lowering move (Growl or Tail Whip). Out of a few dozen battles, the move almost always failed at least once per battle, which is something like a 15% or so chance. I'm a little suspicious that the fail rate is only 1/256.
No, that displays the message "Nothing happened!" in RBY.
Look at any R/B TAS to see what I'm talking about: the Rival uses a stat-lowering move several times in a row, and they all fail. I'm pretty sure those are not all 1/256 misses.
So I'm guessing that applies to all moves that are documented as "having an accuracy"? That would include everything except Swift, Bide, Transform, and non-targeting moves. Thanks!
My second question was that Growl supposedly has "100%" (255/256) accuracy, which means that it should work 99.6% of the time. However, when I've personally seen it used in the opening Rival battle (on a physical Red cartridge), it appears to display "But it failed!" a lot more often than 0.4%. Why?
Under what exact conditions does the 1/256 miss glitch occur?
More specifically, for which of these move categories does the glitch occur?
- Physical moves like Strength
- Special moves like Ice Beam
- Swift
- Bide
- Transform
- Struggle
- Non-attacking moves like Spore
- Non-targeting moves like Amnesia, Reflect, and Haze
I'm asking because I have seen numerous different answers on other websites, and I'd like to get a very clear answer.
On a related note, why, when using a move like Growl in the first rival battle, does it display "But it failed!" much more often than the documented 1/256 miss rate?
"Good goal choice" might be described as "shows off a side or aspect(s) of the game that would not otherwise be seen", such as the SM64 BLJ-less run and the SMW small-only run. I think that a no-coins run could possibly fit this category. It's at least worth a shot.
I never got to 6-1; I was trying to complete the game with warps first so that I could at least save the game any time instead of having to rely on Star Coin gates. But given that 8-T1 seems impossible or at the very least requires inhuman levels of precision... =(
Having done both small-only and tiny-only in realtime (any%) for fun, I'd say the two categories are pretty similar (tiny-only feels slower, though).
I'd also like to suggest minimum-coins as a category (sadly, there are some obviously unavoidable coins in 8-T1, but apart from that, you can even do it warpless without collecting coins in any other level). Playing like that makes NSMB DS feel like a kaizohack rather than a regular Mario game…
I've attempted a no-coins run, and I can tell you that 1-T is a complete ***** to beat, because it requires the Mini Mushroom and several ridiculously precise jumps in a row before the checkpoint. It's worse if you insist that your coin counter stays at 0 the whole run though, because then you can't use the checkpoint and you have to redo the whole Tower again if you pick up a coin. It makes getting a coin significantly worse than dying.
About the unavoidable coins in 8-T1, I know for a fact that many of the walk through walls glitches from SMB3 remain intact in this game, including the one involving big Mario duck-jumping into a corner to walk through it (seen several times in Bowser's Castle). Perhaps, with TAS precision, that could be done to skip the coins in 8-T1? I'm not sure if the mechanics work the same way though.
I think it can be categorised quite simply:
- Does not use BLJs anywhere
- Does not sequence-break the overworld via any method
Vaguely arbitrary, but defined goals. And they certainly lead to an interesting result.
I like these statements.
It might be more specifically defined as:
- Beats the game as fast as possible while also following the below two statements:
- Intentionally sacrifices speed by not using BLJs, for entertainment.
- Intentionally opens the 8 star door, 30 star door, 50 star door, and 70 star door while sacrificing speed to do so, for entertainment (in the last case, "opening" the door means causing the stairs behind it to be non-endless).
For those arguing that one or more of the goals is/are arbitrary, I'm pretty sure the goal of the site has always been entertainment primarily, with speed as a (very strong) secondary goal. If the general audience feels these goals make for an entertaining run, it should get published.
After much thinking, I gave up on trying to define this category under any sort of "do something as fast as possible without doing something else" description, and settled with "aims for entertainment while also getting 70 stars BLJless, with speed as a secondary goal". This kind of puts it in the same vein as the Family Feud and Air (SMB1 Hack) runs.
Voting yes.
This isn't a 100% run; it's an "any% with no BLJs" run.
I'm not sure if this statement is true; I remember it being possible to beat the game with 1 star BLJless by using HSWKs instead.
I unfortunately don't have enough time to watch the whole video or analyze it, but if HSWKs are not abused during the run, then the run goal could be defined as "any% while Mario's speed does not exceed [some number]" where [some number] is whatever is required to get all those wall collision glitches to work. If HSWKs or other speed related glitches are used, then "any%" cannot be used to describe the run, and it should be described as "intentionally gets the fewest number of stars intended by the game developers even though fewer is possible, in order to provide more entertainment".
Can someone tell me why the glitch isn't possible in YI2?
I've followed the main SMW thread for a long time and I still can make absolutely no sense out of this glitch (although I can easily understand the goal sphere and goal tape glitches in the other run).
[quote="dnnzao"I know about the GBA version, but I hate that game, it sucks in so many ways. Plus ISM already did a run for All Yoshi's Coin on GBA :/[/quote]
Any reason why ISM isn't submitting it?
Do not do an all Dragon Coins run on the SNES version. Few levels have them, and it's not even saved after a soft reset.
If you're going to do an all Dragon Coins run, do it on the GBA - it's one of the few things that makes it really unique over the SNES version. The GBA version has 5 Dragon Coins in all 72 (?) main levels except Bowser's Castle, they're saved permanently to the save file, and a special cutscene is displayed upon getting all 360 (?) of them. Also, some of the Dragon Coins, especially in the later levels, are in out-of-the-way areas so such a run would display more of the game.
Does the exit count increase upon using a goal sphere in a Switch Palace?
Specifically what can appear besides the naked koopa and goal sphere? What's PI?
A few questions:
- If you use the goal sphere in Bowser's Castle or any other level without an exit (Yoshi's House, Top Secret Area), does the exit count increase?
- If you use it in a Switch Palace, does the switch activate afterward?
- How do you save fractions of a frame? ("30.5 frames faster")
- Are there any other non-glitchy items that can be obtained in the reserve box?
And just in case, if there was any debate on this earlier:
Using the goal sphere in Bowser's Castle does not count as completing the game, as there is no detectable change to the save file and the credits do not roll.
