Super Metroid - 0:31 SS by Hotarubi!

butterflyboy1974

2008-07-05 05:32:47 pm

Kneel before the great & powerful chOZo

Quote from 072:

How fast would the shoot button need to be pressed for 2-rounding, exactly?

[quote=hero of the day]He had alot of trouble with the missile lake bomb jump, so I suggest practicing the CWJ method instead. It is about 2 seconds faster than the bomb method.

My plan was to go for the CWJ, but revert to DBJing if I knew that I jumped too early. I'm pretty consistent with DBJing.

[quote=hero of the day]Do you plan on making a run with this route? It would be awesome if you could do some test smv's and post them on here.[/quote]

That sounds like a good idea, but I'll need to buy an Snes controller adapter thing first, since I absolutely cannot play on a keyboard.
[/quote]

I got a logitech game pad from walmart for $18 and its better than any nintendo controller I;ve ever had. It even has two extra L & R buttons that I can set to save/reload, turbo fire, etc. i wish I could use it on my consoles.

Cpadolf

2008-07-05 06:05:54 pm

Almost happy

Quote from 072:

How fast would the shoot button need to be pressed for 2-rounding, exactly?

As fast as a 3 round really, which is 6 times per second maybe with a little margin if you run at him instead of stutter walking (while possible the hard part is to do not press a frame to early because then the missile won't be shot until the next press). The thing that is so much harder for a 2 round is that you would probably need to stutter walk which makes precision even tougher. You should train some on it and see how many missiles you can get in at a time.

072

2008-07-05 06:33:25 pm

coral to complement blue

Would it be possible to unleash 10 missiles while running from one side of the room to the other (without holding the run button in) then?

Cpadolf

2008-07-05 08:13:28 pm

Almost happy

Don't think so, although you only need to first 7 (as you can hit him with 3 before he goes of screen then the rest when he comes at you). I still don't know if that is possible, but if you release forward for a short moment just in the middle it might. You'll have to train on this to get a definitive answer.

072

2008-07-05 11:16:37 pm

coral to complement blue

7 doesn't seem so bad.

I'm going to start the test run on the emu then as soon as my adapter arrives. That way I can get feedback on it, plus save myself some frustration practicing Phantoon with savestates.

MASTER-88

Edit history:

MASTER-88: 2008-07-20 09:25:57 am

2008-07-20 09:23:53 am

Speedrunner

I watch that some time ago. This run was totally amazing, especially 2 round Phantoon fight rock Shocked

But i see he doesn,t use murder beam against MB. This can drop 15 seconds game time.

2-round draygon can be done on segmented run, but i just think it is too hard on SS or you need lot luck. There is some mistakes after Botwoon.
Overall this run was amazing job, one greatest run ever in the world. Yes i think Hotarubi can drop at least 1 minutes ingame time. Who really know is 29 minutes SS possible somedays future with Hotarubi skills.

I think only person who can get 29-30 minutes run is Hotarubi. I never believe he can beat his amazing 32 SS run. 31 minutes SS rock. Shocked

Still his 14% single segment run was greatest Super Metroid run ever. His luck and skills like almost TAS. Shocked

His speed was cool and this category is so fucking hard to beat itself even using saves everywhere.

Sneaky Paradox

2008-07-21 04:25:43 am

I've tested a Phantoon first route before and I can that I do not think it is faster. Although I haven't done a typical route speed run since February, I'm pretty darn sure.

Folkskygg

(user is banned)

2008-07-27 09:56:55 am

He's not human! One thing I always wonder is what button setup he/and everyone else who is good chooses. Is it standard because i've never seen somone change the button setup. I change it everytime so I have jump button down(B) boost to the left(Y), fire to the right(A) and weapon select up(X).

MASTER-88

2008-07-27 10:41:04 am

Speedrunner

Quote from Folkskygg:

He's not human!

Really?

I don,t know much about Hotarubi, but i use next controlls:
A-button dash
B-button jump
y-button shots.

I believe this is most common setting with every speedrunner.

Folkskygg

(user is banned)

2008-07-27 11:06:59 am

Quote from MASTER-88:

Quote from Folkskygg:

He's not human!

Really?

I don,t know much about Hotarubi, but i use next controlls:
A-button dash
B-button jump
y-button shots.

I believe this is most common setting with every speedrunner.

I've used that combination also but it's not better nor worse. I had shots on R-button once and that worked pretty good...I had full control(could dash,jump firing at the same time) only problem was tapping really fast.

072

2008-07-27 08:53:05 pm

coral to complement blue

I use:

A - Item Select
B - Jump
X - Shoot
Y - Run

All combinations of x/y/b are possible with just my thumb, and I bring my index finger down for item switch. It's less painful than the "hold run with index finger" way which is what Hotarubi and I think Red Scarlet use/used. The catch with mine is that you can't press select (item cancel) and run at the same time. Only really bothersome in 100% because you have to press item select 4 times instead of 2 to turn off your Supers.

I also switch L and R but that is 'cause I'm weird and have always done that.

rekameohs

2008-09-01 11:25:37 pm

Anywhere, everywhere

Quote from 072:

I use:

A - Item Select
B - Jump
X - Shoot
Y - Run

I also switch L and R but that is 'cause I'm weird and have always done that.

EXACT same here, reversed l and r and everything. That combination lets me jump, shoot, and run with no issues, but switching weapons is a little bothersome. No big deal, though.

Opium

2008-09-01 11:51:41 pm

i use:

A - Fire
B - Jump
X - Item select
Y - Item Cancel
L1 - Dash
R1 - Turbo fire
L2 - diagonal
L3 - diagonal

Having a joypad with an extra set of L & R buttons makes it easier to always hold dash

griggski

2008-10-28 05:19:28 pm

30 minutes! aiwebs_016

Wow, a lot can happen in two years. I used to frequent this board a couple of years ago, and had actually submitted a 36 min SS to SDA...about two weeks before Hotarubi's run came out! That pretty much took the wind out of my sails.
I have always used the default control scheme.

b - dash
a - jump
y - item cancel
x - shoot

I generally shoot with my index finger, and sometimes will use my middle finger to push R. I also reach way over with my right index finger to push select. I've been playing the game so long (bought it in 94' when it came out - I was in 9th grade) that I've just gotten used to that.

Quick question - what is this CWJ that is apparantly being used to clear the missile moat?

Kejardon

2008-10-28 11:22:15 pm

Embarrasing Fact: Power suit made by lowest bidder

Quote from griggski:

Quick question - what is this CWJ that is apparantly being used to clear the missile moat?

Continuous Wall Jump. Samus is actually able to wall-jump off of a surface without turning around, it just requires particularly precise positioning.
There's a relatively easy setup at the missile moat to properly position Samus. First, de-equip the speed booster and high jump boots if you have them, then go to the door that LEADS to the missile moat. Run into the door, THEN open it - this gets Samus at a good starting position. Hold run and go right. In the next room, you have to jump at the last possible frame before getting bogged down into the water. Hold jump up to the height of Samus's jump, then let go to wall jump off of the pillar in the center of the room. You actually have a few frames of room to wall-jump, so the wall jump itself isn't *too* hard to pull off reliably. The tricky part is actually the first jump, but fortunately if you mess that up it doesn't take too long to get back and try again.

[quote="griggski"]
Quick question - what is this CWJ that is apparantly being used to clear the missile moat?
[/quote]
Continuous Wall Jump. Samus is actually able to wall-jump off of a surface without turning around, it just requires particularly precise positioning.
There's a relatively easy setup at the missile moat to properly position Samus. First, de-equip the speed booster and high jump boots if you have them, then go to the door that LEADS to the missile moat. Run into the door, THEN open it - this gets Samus at a good starting position. Hold run and go right. In the next room, you have to jump at the last possible frame before getting bogged down into the water. Hold jump up to the height of Samus's jump, then let go to wall jump off of the pillar in the center of the room. You actually have a few frames of room to wall-jump, so the wall jump itself isn't *too* hard to pull off reliably. The tricky part is actually the first jump, but fortunately if you mess that up it doesn't take too long to get back and try again.

Phazar

2008-10-30 09:28:47 pm

...

My button setup is:

A - Jump

B - Run

X - Shoot

Z - Item Select

This is the default control scheme for the GCN controller.

Smokey

2009-05-09 04:27:43 pm

While I'm not a whining person about games and stuff, I'd like to give my thoughts and experiences that I've had on this for the past 3 years. Since the 0:32 was released, I've been testing exclusively the normal route (no WS first route) on the emulator with savestates only (no slowdown/perfect frame movements). I'm pretty sure any speedrunner has done this, for the obvious reason to predict the outcome of a flawless unassisted run.

After doing many many tests, deep down, I know I will never (or anyone) achieve a 0:32 or sub 0:31. I've done consistently for the past 3 years many 0:32 and lastly only two 0:31. Not only that, but I've done a few 0:32 runs with errors on them to make them look legitimate. To any true vet (at least in the speedrunning community), knows that logic tells that it's possible to improve a time using exactly someone's 'imperfect' run to at least 2 or 3 minutes while using *savestates-only* in a SS; that has been the pattern that I've seen with my skills:

*Using exactly the same tricks and route from my segmented 0:36 - 0:34 is my SS limit.
*Using exactly hotarubi's 0:32 and 0:31 route/tricks - 0:31 SS is my limit

Seems kinda odd to me. I find it impossible to believe someone can perform consistently many tricks in a SS: Kraid's quick-kill, 2 round Phantoon, Draygon 1/2 quick-kill method, pricisely walljumping the 2nd room after Botwoon (although Gravity jump can be used), but also the lack of pressure and cold-nerves to perform them. After seeing the same results, time after time, I became skeptical about his runs and more so after watching a 0:31 (and a 0:44 14%, but that's another matter...)

I made this post because I know some of you (at least people who have broken the sub 0:39 legitimately) know deep down how impossible an amazing 0:32 or 0:31 is.

If, for whatever reason some of you vet speedrunners are skeptical of my point of view I'd greatly like to know your thoughts on this and (if you know) tell me what time you guys have gotten using savestates-only (no slowdowns or frame-by-frame perfect moves). If you have, you gotta have to find it interesting how impossible it is to get 0:32 or below in unassisted conditions. This post is more directly for those of you who have been speedrunning this game long enough.

While I'm not a whining person about games and stuff, I'd like to give my thoughts and experiences that I've had on this for the past 3 years. Since the 0:32 was released, I've been testing exclusively the normal route (no WS first route) on the emulator with savestates only (no slowdown/perfect frame movements). I'm pretty sure any speedrunner has done this, for the obvious reason to predict the outcome of a flawless unassisted run.

After doing many many tests, deep down, I know I will never (or anyone) achieve a 0:32 or sub 0:31. I've done consistently for the past 3 years many 0:32 and lastly only two 0:31. Not only that, but I've done a few 0:32 runs with errors on them to make them look legitimate. To any true vet (at least in the speedrunning community), knows that logic tells that it's possible to improve a time using exactly someone's 'imperfect' run to at least 2 or 3 minutes while using *savestates-only* in a SS; that has been the pattern that I've seen with my skills:

*Using exactly the same tricks and route from my segmented 0:36   -  0:34 is my SS limit.
*Using exactly hotarubi's 0:32 and 0:31 route/tricks  -    0:31 SS is my limit

Seems kinda odd to me. I find it impossible to believe someone can perform consistently many tricks in a SS: Kraid's quick-kill, 2 round Phantoon, Draygon 1/2 quick-kill method, pricisely walljumping the 2nd room after Botwoon (although Gravity jump can be used), but also the lack of pressure and cold-nerves to perform them. After seeing the same results, time after time, I became skeptical about his runs and more so after watching a 0:31 (and a 0:44 14%, but that's another matter...)

I made this post because I know some of you (at least people who have broken the sub 0:39 legitimately) know deep down how impossible an amazing 0:32 or 0:31 is.

If, for whatever reason some of you vet speedrunners are skeptical of my point of view I'd greatly like to know your thoughts on this and (if you know) tell me what time you guys have gotten using savestates-only (no  slowdowns or frame-by-frame perfect moves). If you have, you gotta have to find it interesting how impossible it is to get 0:32 or below in unassisted conditions. This post is more directly for those of you who have been speedrunning this game long enough.

Saturn

2009-05-10 10:25:50 am

I can clearly understand you being skeptic about this, Smokey. Seeing that even a highly skilled player like you has such doubts about hotarubi's runs being legit, while giving pretty accurate examples and reasons to them, this can indeed be something to take a closer look at from the speedrunning community.

I actually felt exactly the same way when I saw his any%, and especially the crazy 14% SS run, always asking myself how a player could be that cool and overcome even the hardest and most critical situations without screwing up in them for such a long time. As a average player who managed to get 0:37 SS, I totally agree that reaching a 0:31 SS in unassisted conditions seems to be far beyond possible.

But seeing that, according to nate, his 0:32 got verified by SDA for a very long time (and they should be experienced enough to spot a fake run, after all the previous attempts where they succeeded with this as well), I actually believe, as hard as it is, that his runs must be legit. My only justification for this is that hotarubi just has some very rare, and extraordinary talent.

In the end though, it will probably never be possible to know for sure, unless you'd see the player performing such a run in front of your eyes.

I can clearly understand you being skeptic about this, Smokey. Seeing that even a highly skilled player like you has such doubts about hotarubi's runs being legit, while giving pretty accurate examples and reasons to them, this can indeed be something to take a closer look at from the speedrunning community.

I actually felt exactly the same way when I saw his any%, and especially the crazy 14% SS run, always asking myself how a player could be that cool and overcome even the hardest and most critical situations without screwing up in them for such a long time. As a average player who managed to get 0:37 SS, I totally agree that reaching a 0:31 SS in unassisted conditions seems to be far beyond possible.

But seeing that, according to nate, his 0:32 got verified by SDA for a very long time (and they should be experienced enough to spot a fake run, after all the previous attempts where they succeeded with this as well), I actually believe, as hard as it is, that his runs must be legit. My only justification for this is that hotarubi just has some very rare, and extraordinary talent.

In the end though, it will probably never be possible to know for sure, unless you'd see the player performing such a run in front of your eyes.

Smokey

Edit history:

Smokey: 2009-05-23 03:19:08 pm

2009-05-10 03:41:21 pm

I completely agree with you, Saturn. I know people wouldn't take my word alone, but if more experienced people (who have gotten below 0:39 at least) come and analyze his runs will notice a peculiar behavior on them. By the way, I'm not saying that Nate (or Radix) didn't do a good job verifying them (I really don't know what procedures they took). But when you equip yourself with the aid from very experienced people (again, people who can get below 0:39/legit) you'll definitely spot something not just in quaility, but in general skill behaivor and movements.

Take for instance the fact that ALL of his runs are SS. Any speedrunner knows that segmented runs tends to improve your SS time. If so, why doesn't he do one segmented to see if he can lower 0:31?... Another example is his SS 14%. Seriously, 0:44? Not even before he fought Ridley he saved.

As for movements, I've been *speedrunning* using the keyboard for a very long time to see moves in common. Why is it that every time he uses the charge beam while dashing, he arm-pumps with just one shoulder button instead of both, and after a few seconds of running, he begins to press both in some cases? Not just one time, he does it almost every time. That's because sometimes you can't toggle to many buttons at once.(<I certainly do this while using the original Zsnes keyboard setup)

If anybody cares or want to see, I uploaded one of my smvs, thanks to Cpadolf. It's my latest one and achieves a time of 0:31 SS. Again, I used savestates and rerecords. It doesn't mean to represent a complete professional TAS. I only savestated in every door transition and played the rest of the rooms normally with the best of my ability. I could probably have done 0:30, but they only way was if I had a better ammo management before entering Tourian.

I did another one after this one and the only place I improved was skipping the recharge stop at the ship before Tourian. Guess what, I got again 0:31.

Again, don't take my word from this alone; I'd really appreciate if somebody has the time to produce a savestate-only run for comparison purposes to discuss them. All in all, maybe he indeed has a unique talent. But it isn't certainly the one to get you this very far in unassisted conditions, at least IMO.

Thanks for the reply, Saturn.

EDIT:

Sorry, I forgot the link: http://dehacked.2y.net/microstorage.php/info/709328167/Super_MetroidFINAL.smv

I completely agree with you, Saturn. I know people wouldn't take my word alone, but if more experienced people (who have gotten below 0:39 at least) come and analyze his runs will notice a peculiar behavior on them. By the way, I'm not saying that Nate (or Radix) didn't do a good job verifying them (I really don't know what procedures they took). But when you equip yourself with the aid from very experienced people (again, people who can get below 0:39/legit) you'll definitely spot something not just in quaility, but in general skill behaivor and movements.

Take for instance the fact that ALL of his runs are SS. Any speedrunner knows that segmented runs tends to improve your SS time. If so, why doesn't he do one segmented to see if he can lower 0:31?... Another example is his SS 14%. Seriously, 0:44? Not even before he fought Ridley he saved.

As for movements, I've been *speedrunning* using the keyboard for a very long time to see moves in common. Why is it that every time he uses the charge beam while dashing, he arm-pumps with just one shoulder button instead of both, and after a few seconds of running, he begins to press both in some cases? Not just one time, he does it almost every time. That's because sometimes you can't toggle to many buttons at once.(<I certainly do this while using the original Zsnes keyboard setup)

If anybody cares or want to see, I uploaded one of my smvs, thanks to Cpadolf. It's my latest one and achieves a time of 0:31 SS. Again, I used savestates and rerecords. It doesn't mean to represent a complete professional TAS. I only savestated in every door transition and played the rest of the rooms normally with the best of my ability. I could probably have done 0:30, but they only way was if I had a better ammo management before entering Tourian.

I did another one after this one and the only place I improved was skipping the recharge stop at the ship before Tourian. Guess what, I got again 0:31.

Again, don't take my word from this alone; I'd really appreciate if somebody has the time to produce a savestate-only run for comparison purposes to discuss them. All in all, maybe he indeed has a unique talent. But it isn't certainly the one to get you this very far in unassisted conditions, at least IMO.

Thanks for the reply, Saturn.

EDIT:

Sorry, I forgot the link: http://dehacked.2y.net/microstorage.php/info/709328167/Super_MetroidFINAL.smv

Smokey

2009-05-10 07:59:05 pm

Here is another one made with some intentional errors to make it look more legit.

http://dehacked.2y.net/microstorage.php/info/1217427951/Super_Metroid2.smv
Final time: 0:32

I hope Nate sees this. Not just him, but I really hope all the SM vets see this as well. Maybe in the end, more people will take a closer look at the current 0:32 and give it a second thought.

nate

2009-05-11 01:20:51 am

yeah, i've been watching this. sort of been waiting to weigh in. i should say that what follows is my personal opinion and has nothing to do with sda since i have nothing to do with the verification process.

imo it's a question of whether 100% of the run is possible. in my experience it's very hard to make something perfectly possible and excellent at the same time. it will end up uneven to some degree, with some rooms too perfect and some mistakes too "pre-meditated." and at least one of the joints will show.

of course what is considered possible changes all the time. for example your runs underwent the same highly critical process before they were posted on sda because some people felt it was highly probable you were cheating. this was before modern sda verification so it was a very special case. but no evidence of cheating was ever found, same as with hotarubi's stuff.

sda basically operates on the principle that it's more work to make a "perfectly possible and yet excellent at the same time" run than to just run it without cheating. something will inevitably go wrong; there will be something in there like bob in fusion that you can find that will show it is cheated. but nothing like that has ever been found in hotarubi's runs to my knowledge.

about the keyboard stuff, sda has occasionally requested video proof of control ability, especially in cases of rapid fire. having seen frezy_man play contra in person last january i have expanded my view of what a human body is capable of doing quite a bit. the speedrunning community is still rather small and so it's possible to get new people who really redefine the game as it were. i want to say that i looked at and listened to his stuff every time it came to me to encode and i never found any evidence it was emulated. of course that doesn't mean he's not playing with a keyboard but it does decrease the likelihood any of it is cheated. kejardon may want to talk about what he saw because i know he was consulted on at least one of the runs.

yeah, i've been watching this. sort of been waiting to weigh in. i should say that what follows is my personal opinion and has nothing to do with sda since i have nothing to do with the verification process.

imo it's a question of whether 100% of the run is possible. in my experience it's very hard to make something perfectly possible and excellent at the same time. it will end up uneven to some degree, with some rooms too perfect and some mistakes too "pre-meditated." and at least one of the joints will show.

of course what is considered possible changes all the time. for example your runs underwent the same highly critical process before they were posted on sda because some people felt it was highly probable you were cheating. this was before modern sda verification so it was a very special case. but no evidence of cheating was ever found, same as with hotarubi's stuff.

sda basically operates on the principle that it's more work to make a "perfectly possible and yet excellent at the same time" run than to just run it without cheating. something will inevitably go wrong; there will be something in there like bob in fusion that you can find that will show it is cheated. but nothing like that has ever been found in hotarubi's runs to my knowledge.

about the keyboard stuff, sda has occasionally requested video proof of control ability, especially in cases of rapid fire. having seen frezy_man play contra in person last january i have expanded my view of what a human body is capable of doing quite a bit. the speedrunning community is still rather small and so it's possible to get new people who really redefine the game as it were. i want to say that i looked at and listened to his stuff every time it came to me to encode and i never found any evidence it was emulated. of course that doesn't mean he's not playing with a keyboard but it does decrease the likelihood any of it is cheated. kejardon may want to talk about what he saw because i know he was consulted on at least one of the runs.

Saturn

Edit history:

Saturn: 2009-05-11 11:52:00 am

2009-05-11 11:43:47 am

This are indeed some good points that could indicate that a emulator, or at least a keyboard was used. But as said, I don't want to make wrong assumptions about hotarubi without having concrete proof.

As for why he used SS, probably because every save and reload between segments costs you ~8 seconds, which is pretty decent for a quality speedrun. I guess he just managed to get consistent enough to get a better result from a SS overall, by eliminating this delays. Although I agree that in a 14% run it would in any case be more benefitting to use at least one save before Ridley, as that fight is indeed one of the most risky and difficult tasks you can get in any kind of run in this game. Maybe he just wanted to get 14% SS because of the awesome accomplishment it is in itself, regardless of the time.

As for your runs, Smokey, I watched both of them. Great skills without a doubt (I recognized your crazy walljump-patterns right away ;-)), considering you used savestates between door transitions only. I got curious what I could come up with trying to do the same, so I went ahead and finished one as well. However I wanted to include every realistic speed trick in it, so I had to use a 50% slowdown on 2 spots, to execute fast enough walljumps for the pre-Draygon shinespark, and to do the speedball through the bomb blocks in the spike/kago room of LN when returning back after Ridley. Other than that I did everything in realtime, but had to use savestates more often at critical parts for that, not only between door transitions, as I'm just not that skilled/consistent at such a high speed level. Furthermore I used them to get good refill drops, to simulate optimal luck in a speedrun.

With all this, I managed to get a mid 0:31 as well. The route is basicly the same, with the only difference that I skipped the 3rd Super Missile pack, and the refills on the sand bugs after Botwoon, because I believe that the delays of them wouldn't make up for the gains at Draygon, considering he can be easily 3-rounded with shinesparks only, while at Ridley, you already have the strong Plasma beam. Oh, and the ship refill before Tourian is in any case a huge time waste and never worth it IMO. Even without luck manipulation, the Metroids should drop enough Super Missile refills to compensate the ship refill (you can always use Missiles to kill them if it gets close), and the energy is not a concern, since you have to use the refill station after the big Metroid anyway.

I'm pretty sure I could get 0:30 with this runs strategies if playing on Smokey's level, but I'm just not that skilled. Even if I would skip the pre-Draygon shinespark and the LN exit speedball where I was forced to use a slowdown, I would most likely still end up with a high 0:31, having a margin of ~25 seconds of errors in that run, which is overall a much better result than I thought would be reachable with my realtime-skill level. I think this clearly shows how important consistency is and that skill is not the only factor that is required to get a top speedrunner. :-)

Anyway, here is the run for those interested. Recorded on Snes9x 1.43 v12, but should work on any of the later versions of the 1.43 series. Enjoy!

EDIT:
You have to copy-paste the emulator link in your browser, since for some reason it doesn't open the link if clicking on it...

This are indeed some good points that could indicate that a emulator, or at least a keyboard was used. But as said, I don't want to make wrong assumptions about hotarubi without having concrete proof.

As for why he used SS, probably because every save and reload between segments costs you ~8 seconds, which is pretty decent for a quality speedrun. I guess he just managed to get consistent enough to get a better result from a SS overall, by eliminating this delays. Although I agree that in a 14% run it would in any case be more benefitting to use at least one save before Ridley, as that fight is indeed one of the most risky and difficult tasks you can get in any kind of run in this game. Maybe he just wanted to get 14% SS because of the awesome accomplishment it is in itself, regardless of the time.

As for your runs, Smokey, I watched both of them. Great skills without a doubt (I recognized your crazy walljump-patterns right away ;-)), considering you used savestates between door transitions only. I got curious what I could come up with trying to do the same, so I went ahead and finished one as well. However I wanted to include every realistic speed trick in it, so I had to use a 50% slowdown on 2 spots, to execute fast enough walljumps for the pre-Draygon shinespark, and to do the speedball through the bomb blocks in the spike/kago room of LN when returning back after Ridley. Other than that I did everything in realtime, but had to use savestates more often at critical parts for that, not only between door transitions, as I'm just not that skilled/consistent at such a high speed level. Furthermore I used them to get good refill drops, to simulate optimal luck in a speedrun.

With all this, I managed to get a mid 0:31 as well. The route is basicly the same, with the only difference that I skipped the 3rd Super Missile pack, and the refills on the sand bugs after Botwoon, because I believe that the delays of them wouldn't make up for the gains at Draygon, considering he can be easily 3-rounded with shinesparks only, while at Ridley, you already have the strong Plasma beam. Oh, and the ship refill before Tourian is in any case a huge time waste and never worth it IMO. Even without luck manipulation, the Metroids should drop enough Super Missile refills to compensate the ship refill (you can always use Missiles to kill them if it gets close), and the energy is not a concern, since you have to use the refill station after the big Metroid anyway.

I'm pretty sure I could get 0:30 with this runs strategies if playing on Smokey's level, but I'm just not that skilled. Even if I would skip the pre-Draygon shinespark and the LN exit speedball where I was forced to use a slowdown, I would most likely still end up with a high 0:31, having a margin of ~25 seconds of errors in that run, which is overall a much better result than I thought would be reachable with my realtime-skill level. I think this clearly shows how important consistency is and that skill is not the only factor that is required to get a top speedrunner. :-)

Anyway, here is the run for those interested. Recorded on [url=http://www.freewebs.com/saturnsmovies/snes9x-1.43-improvement12-win32.zip]Snes9x 1.43 v12[/url], but should work on any of the later versions of the 1.43 series. Enjoy!

EDIT:
You have to copy-paste the emulator link in your browser, since for some reason it doesn't open the link if clicking on it...

attachment:

Saturn - SM in 31 - savestate run v1.zip

072

2009-05-11 11:54:58 am

coral to complement blue

As for why he always does SS -- remember that many people consider segmented runs to be "impure"...

Guardian Garr

2009-05-11 11:17:33 pm

I didn't read all the comments, but Cpadolf, you seem to be surprised by just about anything lol. As far as I am concerned, if hasn't been proven yet, it is possible.

Cpadolf

Edit history:

Cpadolf: 2009-05-12 06:52:02 am

2009-05-12 03:39:13 am

Almost happy

What the hell do I have to do about this? I made no comment here and I do believe hotarubi's run is legit and possible to recreate.

I get surprised when people say they can do things I think of as impossible, and in those cases I demand good proof in order to believe it. And so far I think my suspicions has been fair, seeing as nothing I have claimed impossible has yet been proven possible to my knowledge.

EDIT: In unassisted play at least...