Super Mario World 11-exit Improvement

Lucid Faia

2007-04-03 01:19:27 am

Yeah, I said it.

Quote:

looks like you got something to beat lucid. Smiley

Quote:

Go lucid, go! Tongue

Probably he had no SNES. but if you play and beat a record for a hole summer you can buy a SNES. Maybe he isn´t as serious as nerds here. Maybe he was very poor or don´t care very much about rules and stuff.

I had already done about 50 attempts before these posts. =)

I'm sure it was easy to figure out, but the rejected submission that was "barely an improvement" was mine. It was done at a time when simply "getting the record" was more important to me than truly optimizing the run and going as fast as possible, and I think that's why it was rejected. Perhaps, rightfully so. That, combined with the archive.org run, has motivated me to do a massive improvement.

Meh. Casual gaming and speedrunning do not mix. I should be playing Guitar Hero II or God of War II, and instead I'm playing Metal Gear Solid 2 and Super Mario World. Damn you, SDA. >_<

andrewg

Edit history:

andrewg: 2007-04-03 01:45:05 am

2007-04-03 01:43:42 am

Hi! I'm andrewg!

i think I'm going to give this another shot. I'd like to at least improve my crappy 12:14. :-/

Maybe I'll get addicted to this like I did smb1.

stanski

2007-04-03 02:22:00 am

wise fwom yo gwave

The video quality was very bad for such a small improvement. I think if it had been a less popular game, it might have been ok, but that is one of the first runs people see when they get to the site, and you don't want to have something that blinds the person be the first impression the site gives.

Good luck on improving it :-)

skamastaG

2007-04-03 02:34:04 am

I desperately hope one of you guys beats the record, so I will be stongly motivated to get a perfect time.
The only thing in that xbox run that impressed me was donut secret house and the bowser fight. Both of those I have barely practiced but I'm pretty sure thegreginator can beat/tie both of those times.
That run can be improved by at least 5 seconds if not more. I am way more devoted to a smw TAS I'm working on right now but I will try to run this every once in a while.

As for the 96-exit record I said I wanted to beat a while back, I have such a short attention span when trying to do full runs that I usually get bored and make a stupid mistake and just stop the run. I also rarely want to dedicate one and a half hours to do a speedrun so 11-exit suits me much better.

DK64_MASTER

2007-04-03 08:17:35 pm

Quote:

I also rarely want to dedicate one and a half hours to do a speedrun so 11-exit suits me much better.

You need an hour and 21 minutes not 1 and a half hours. Wink

SqueamishCow

2007-04-03 09:40:10 pm

I think you mean 1 hour and 31 minutes.

andrewg

Edit history:

andrewg: 2007-04-04 01:40:57 am

2007-04-04 01:39:30 am

Hi! I'm andrewg!

So i've been practicing the levels. This seems easier than super mario bros. 1 to speedrun, but I haven't quite tried the later levels too much for speedrunning.

The progress goes well though. I've been trying to see how fast I can do both the first caslte and the level where you get the cape. I seem to have little to no success so far. :-/

what is your game unit time ending the level where you get the cape?

skamastaG

2007-04-04 05:24:13 am

361 is a good time for donut plains 1(level where you get cape)

362 is the best we have been able to get, but I think you can just barely get a 363 with perfect flight speed.

DK64_MASTER

Edit history:

DK64_MASTER: 2007-04-04 10:11:34 am

2007-04-04 10:10:59 am

Quote:

I think you mean 1 hour and 31 minutes.

Nope, the TAS is 1 hr 21 minutes. skamastaG can beat that. (notice the Wink

at the end).

You better beat it!

Lucid Faia

2007-04-04 04:38:46 pm

Yeah, I said it.

In general, Super Mario World is definitely more complex and harder to speedrun than Super Mario Bros. It just seems like SMB is harder because of the demanding standard that has been set for it. We're a long way from the SMW equivalent of sdkess's SMB run.

I haven't utilized any "oscillating speed" tricks, so I'll be looking for standard times of 360 on YI2 and 360 on DP1, etc.

I'm not a fan of duck-jumping through the ghost ring on DSH. It's actually not that easy, it only saves one time unit, and the "big three" of DS1, DSH, and SW1 are hard enough levels to begin with. The best time I've gotten without using this tactic is 338, but I think I can do better than that. I'll run some more tests later tonight.

I'm also trying that Smashola method for SW1. I understand how it works pretty well, but it doesn't seem to be reliable. It looks like I'm emulating the TAS movements very well, but I'm only triggering the glitch about half the time. I need to practice this.

andrewg

Edit history:

andrewg: 2007-04-06 06:21:42 pm

2007-04-04 08:58:23 pm

Hi! I'm andrewg!

"In general, Super Mario World is definitely more complex and harder to speedrun than Super Mario Bros. It just seems like SMB is harder because of the demanding standard that has been set for it. We're a long way from the SMW equivalent of sdkess's SMB run. "

yes it's more complex. it is easier due to being perfect on every level in smb1. I mean, I can make mistakes in the smw run and still be good, which means it's easier. The game itself is harder though.

This has been pretty tough though. especially the castles. I think after I do super mairo bros. 1 I'm going to really concentrate on smw. Although I feel as if the smb2 run needs some improving.

Lucid Faia

2007-04-13 05:15:33 pm

Yeah, I said it.

Just thought I'd post this here and ask what people think of this. The 11:09 run I posted in this topic, and an improvement I recently submitted, were both rejected for not being big enough improvements over sdkess's run. However, I just found out that this is due to how SDA times SMW runs. I don't know if it's fair or not to lobby for a rule change (in fact, despite my feelings on the matter, it probably isn't), but I do want to see if others think this makes any sense.

I've been told that on SDA, timing for SMW ends when Bowser flips upside-down and Princess Toadstool appears. Here's the problem; after being hit for the sixth and final time, Bowser has to rise to the top of the screen before he can flip like this. In sdkess's run, Bowser was hit at the top of the screen, and no extra movement was required. In my latest run, Bowser is much closer to the ground, and takes a couple extra seconds to rise to the top of the screen. Effectively, my run was over two seconds faster when Bowser was hit, but wound up being almost identical to sdkess's time because of this timing rule.

The argument made for timing things this way is that it's when you lose control of Mario. However, when you hit Bowser for the sixth time, all Mechakoopas on the screen disappear and you lose the ability to pause. It's impossible to die, pause, or lose any more time in any way; the ending sequence is technically underway. Really, you lose control of the game when Bowser is hit for the last time. Running and jumping around after that point are meaningless flourishes that don't affect anything. This isn't like Mega Man X where after the final battle, you have to manually close dialogue boxes for the game to progress. Once Bowser is hit, the ending sequence is in progress and you can't pause it or slow it down even if you want to.

What if we were to apply this rule to other Mario games? What if timing in SMB ended when Princess Toadstool appears, rather than when the final axe is touched? Imagine someone clearing the game faster than anyone else has yet, only for it to be rejected because Bowser was in the air when the axe was touched, and took longer to fall to his death. That's basically the equivalent of what has happened here.

So, while I'm not really asking that the rule be changed (at least, not anymore), I do want to see if timing SMW this way makes sense to anyone else. Because to me, it sure doesn't.

Just thought I'd post this here and ask what people think of this. The 11:09 run I posted in this topic, and an improvement I recently submitted, were both rejected for not being big enough improvements over sdkess's run. However, I just found out that this is due to how SDA times SMW runs. I don't know if it's fair or not to lobby for a rule change (in fact, despite my feelings on the matter, it probably isn't), but I do want to see if others think this makes any sense.

I've been told that on SDA, timing for SMW ends when Bowser flips upside-down and Princess Toadstool appears. Here's the problem; after being hit for the sixth and final time, Bowser has to rise to the top of the screen before he can flip like this. In sdkess's run, Bowser was hit at the top of the screen, and no extra movement was required. In my latest run, Bowser is much closer to the ground, and takes a couple extra seconds to rise to the top of the screen. Effectively, my run was over two seconds faster when Bowser was hit, but wound up being almost identical to sdkess's time because of this timing rule.

The argument made for timing things this way is that it's when you lose control of Mario. However, when you hit Bowser for the sixth time, all Mechakoopas on the screen disappear and you lose the ability to pause. It's impossible to die, pause, or lose any more time in any way; the ending sequence is technically underway. Really, you lose control of the [b]game[/b] when Bowser is hit for the last time. Running and jumping around after that point are meaningless flourishes that don't affect anything. This isn't like Mega Man X where after the final battle, you have to manually close dialogue boxes for the game to progress. Once Bowser is hit, the ending sequence is in progress and you can't pause it or slow it down even if you want to.

What if we were to apply this rule to other Mario games? What if timing in SMB ended when Princess Toadstool appears, rather than when the final axe is touched? Imagine someone clearing the game faster than anyone else has yet, only for it to be rejected because Bowser was in the air when the axe was touched, and took longer to fall to his death. That's basically the equivalent of what has happened here.

So, while I'm not really asking that the rule be changed (at least, not anymore), I do want to see if timing SMW this way makes sense to anyone else. Because to me, it sure doesn't.

nate

2007-04-13 05:22:48 pm

it is logical. however i think it may be problematic to phrase for the rules. what would you call end of control now? end of "meaningful" control? wtf is "meaningful"? definitely a problematic case. only chance is to put a timing note on the page itself i think.

InsipidMuckyWater

Edit history:

InsipidMuckyWater: 2007-04-13 05:29:40 pm

2007-04-13 05:28:42 pm

Visit my profile to see my runs!

End of 'effectual' control? Or 'relevant' control? That is, the end of control which could still have relevance to the progression of the game.

Those make sense to me; isn't there a precedent for this kind of ruling with God of War? I can't recall at the moment.

Lucid Faia

Edit history:

Lucid Faia: 2007-04-13 05:39:41 pm

2007-04-13 05:36:26 pm

Yeah, I said it.

Indeed, with most games, timing ends when the player loses control because that's when the game is "won" and the player can't change that in any way, nor can the player halt the process by pausing. With SMW, the game is "won" when Bowser is hit for the sixth time. Being able to control Mario for a couple seconds after that doesn't make a bit of difference in what happens afterwards.

Quote:

only chance is to put a timing note on the page itself i think.

That would be fine. Really, I'm just annoyed that I thought I had the run beat (in fact, it would be considered a three-second improvement on a certain other site), only to find out that I haven't because of this timing rule. I'm not criticizing anyone personally for any of this, but I'm sure you can understand how perturbed I am.

nate

2007-04-13 05:46:53 pm

absolutely. it jumped out of nowhere, one of those rare monkey wrenches. not much going on atm but tonight hopefully it'll get discussed.

stanski

2007-04-13 08:04:15 pm

wise fwom yo gwave

I think this applies to a lot of bosses in games where you "can't effect them but they take longer to die," except you can.

For example, if you hold off for the last hit in sonic 1's final boss, he takes a shorter time to explode. You can say "but you can affect that!" but you can also effect when you land the final hit on the boss in SMW.

I can give more examples if need be as well, but if you think about it, a lot of games have this.

Lucid Faia

2007-04-13 08:53:43 pm

Yeah, I said it.

Quote:

I think this applies to a lot of bosses in games where you "can't effect them but they take longer to die," except you can.

For example, if you hold off for the last hit in sonic 1's final boss, he takes a shorter time to explode. You can say "but you can affect that!" but you can also effect when you land the final hit on the boss in SMW.

I can give more examples if need be as well, but if you think about it, a lot of games have this.

True, but I don't think that's the issue here. The issue is when timing should end. If I had known what the timing rules were, I would have brought this up a long time ago. In my opinion, ending timing when Toadstool appears is silly.

Floogal

Edit history:

Floogal: 2007-04-13 09:05:15 pm

2007-04-13 09:04:21 pm

This brings to mind Kirby's Adventure for the NES (note that there is currently no run for this at the moment).

Once you deliver the final hit to the final boss, there is a good 10-15 seconds where you can still control Kirby, but you can't die. Now, where the boss is on the screen when you defeat him doesn't have an impact (unlike Super Mario World), but there is the fact that if you rapidly shoot starts across the screen, you'll create a lag that can extend that 10-15 seconds by almost 100%. So should we ignore that the player can create lag after the final boss hit, where the game is effectively "won," or wait until after the white screen fades to black?

The point of this post is that each game can have unique rules, and each runner could have different interpretations of the rules. Someone thinking that the game ended on the final hit would be ticked-off to find that the lag s/he creates afterwards (for entertainment purposes) counted against him/her.

So perhaps each game page should explicitly say when timing starts and ends. Due to the overhead involved in doing this, maybe you could email each runner and get them to send in a description.

Although, they weren't the ones doing the timing....

Hmm....

This brings to mind Kirby's Adventure for the NES (note that there is currently no run for this at the moment).

Once you deliver the final hit to the final boss, there is a good 10-15 seconds where you can still control Kirby, but you can't die.  Now, where the boss is on the screen when you defeat him doesn't have an impact (unlike Super Mario World), but there is the fact that if you rapidly shoot starts across the screen, you'll create a lag that can extend that 10-15 seconds by almost 100%.  So should we ignore that the player can create lag after the final boss hit, where the game is effectively "won," or wait until after the white screen fades to black?

The point of this post is that each game can have unique rules, and each runner could have different interpretations of the rules.  Someone thinking that the game ended on the final hit would be ticked-off to find that the lag s/he creates afterwards (for entertainment purposes) counted against him/her.

So perhaps each game page should explicitly say when timing starts and ends.  Due to the overhead involved in doing this, maybe you could email each runner and get them to send in a description.

Although, they weren't the ones doing the timing....

Hmm....

stanski

Edit history:

stanski: 2007-04-13 09:21:52 pm

2007-04-13 09:21:04 pm

wise fwom yo gwave

Well, my point was that timing doesn't end after the final blow is delivered in that game (sonic 1) either, it ends a few seconds later, even though you can't affect the timing after the final blow is delivered. It explicitly states how timing is done on sda's front page, I don't really see how its ambiguous at all. I realize it really sucks that you should have the record twice now, but maybe you can create an even better run now?

Easiest answer to kirby is don't create lag.

snorlax

2007-04-13 09:54:02 pm

zzzzzzzzzz

Doesn't timing in aLttP end on the last hit on Ganon, despite the fact that you can still jump into the pit after defeating him? That seems like a fairly good precedent.

Lucid Faia

2007-04-13 11:22:30 pm

Yeah, I said it.

Quote:

I realize it really sucks that you should have the record twice now, but maybe you can create an even better run now?

Nah, I was borrowing an SNES from my sister, and I've already returned it. I'll just send the run to Twin Galaxies, and maybe put it on Youtube or something. Besides, I recently decided I wouldn't even play another game until I finished the MGS2 run I'm working on.

Radix

2007-04-14 12:20:21 am

I'm addicted to games

Quote:

Doesn't timing in aLttP end on the last hit on Ganon

no

Lucid Faia

Edit history:

Lucid Faia: 2007-04-14 01:37:36 am

2007-04-14 01:33:26 am

Yeah, I said it.

By the way, if SDA is imposing a "one-size-fits-all" rule for timing games, why doesn't SMW timing begin after pressing Start to select "1-Player Game?" The first moment you control Mario is not on the world map, but right before the "Welcome to Dinosaur Land" message box appears. Not only do you have to manually close the message box, but Mario can even briefly run and jump. Why is that meaningless character movement so important at the end of the game, but not at the beginning?

Fabian

2007-04-14 02:51:50 am

I love YaBB 1G - SP1!

In my opinion, what makes most sense for a console run is to stop timing after Bowser takes his sixth hit. This is where I consider the game to be beaten, and whether or not Mario can be controlled or not is completely irrelevant in my opinion. In tool assisted SMW speedruns, we usually end timing immediately after Mario has thrown the last dude, as long as it's set up in a way so Bowser still gets hit on the first possible frame. Something like that doesn't sound very practical or even possible for a console run though, which is why I think the last hit on Bowser is where the game/timing ends/should end.

Just adding my opinion.