Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
page  <- 1234567891011121314151617 -> <- 1 .. 10 .. 17 ->
--
--
List results:
Search options:
Use \ before commas in usernames
Edit history:
Omnigamer: 2014-08-29 08:30:30 am
Omnigamer: 2014-08-29 08:30:14 am
All the things
I think they cared about it mainly from an arbitrary benchmarking perspective. Pretty much the same thing as "I have to cut the 30!!" in any other game; a goal that they know is achievable, but is not necessarily anything special about it other than it being a number. If they really wanted to pursue IGT, there are a number of optimizations and things they would have to go back to, such as Murder Beam. The "traditional" route (used in Zoast's 44:08) just happens to be fairly good for both real time and game time, though apparently not ideal for either.
Edit history:
Reeve: 2014-08-30 06:30:35 pm
Reeve: 2014-08-30 06:27:51 pm
Reeve: 2014-08-30 10:36:29 am
Reeve: 2014-08-30 06:10:58 am
Reeve: 2014-08-29 08:25:26 pm
Reeve: 2014-08-29 08:24:38 pm
Reeve: 2014-08-29 08:17:22 pm
Reeve: 2014-08-29 08:12:34 pm
Reeve: 2014-08-29 08:12:16 pm
I'd like to make a few comments about Behemoth's run myself. Maybe I can help clearing some doubts.

Crateria Shaft Climb: Although it's a slower strategy by around 1 sec Behemoth feels safer doing that than the walljumps and in this game you sometimes go for safer strategies for a greater good. I know he delivers crazy shit risky strategies all around, but in this particular situation the ledge grabs is what he feels more confortable with.

WS Lake: A run is not doomed if you miss the perfect jump to set the cwj animation as long as you can bomb jump to the other side. It's a good idea to practice that strategy as a backup plan and in previous runs that didn't reach sub 44 behemoth had to resort to this artifice. A bomb jump there actually looks really cool.

Painfull Mistake: The mistake after WS Lake where Samus fell in the water costed around 8 seconds and, afaik, was the biggest mistake in the run. Had it to happen several times when practicing in that same spot. Guess it's the jump that needs the most of your attention in that room.

Super Missile Room After Phantoon: Behemoth usually jumps straight to the Super Pack platform and damage boosts back to the door. Thing is he missed the timing for the jump to the Super and samus fell off the platform which caused a not expected damage. Another damage there would make things harder.

Dive in the water after Gravity Suit: That path is not intended at all as you guys probably know. The idea was to damage boost off of the flyer and land on the platform over the water but he pressed back too early and just went through the flyer.

Three Crabs Before Ice: The strategy he wanted there was to emerge in the other side, unmorph, shoot one of the crabs (there would be only two remaining after shooting before morphing to the small passage) open the door and damage boost to Ice Beam room. He tried to unmorph too early and got trapped there. Then he had to kill all crabs.

After HI-Jump: The strategy there is to jump through the HI-Jump door and softmorph in the next room. Then just stand up and walljump up landing a Power Bomb in mid air. As he m issed the jump through the door he felt like the crab would be in the way when trying to walljump up and decided to kill it for safety. Unfortunatelly he ended up messing the room a bit. If done correctly the under Missile Pack route can be around 20 frames faster if you don't count the Missile Pack Box.

Bubble Room: You can notice behemoth shoots up when jumping. By doing that he intended to freeze the flyer right above, then freeze the second flyer by walljumping off of the left door and shooting diagonally up making a kind of stair to the top, kill the cacti to get it's drop and jump to the green door. This strategy is almost 1 second faster than the usuall walljump with pseudo screw.

Bugs Pipe farming: In Behemoth's first attemps he would stay there till he had almost full ammo and health because it's crucial for LN and mainly Ridley. It would take around 7 round in average iirc but the bugs can be very unfair sometimes not giving anything. Having 9 Supers for Ridley saves much more time than you can first imagine. After analyzing a bit more we found out that playing safe from WS to the Bugs Pipe would be the best option. That's why you see Behemoth using missiles in Pinky doors even when he has lots of supers. You can notice how confortable he was after SpeedBooster in terms of Supers and he only needed to stay in the pipe for three waves.

LN Spark: Very tricky but waaaay easier than the spark up to Plasma in Maridia. Behemoth used to gravity jump up that area which would bring 2 more pause screens to the run. This trick eliminates the pauses and saves around 4 and half seconds.

Ridley's Energy Tank: That came to the run after we noticed the need of farming health to survive LN. Getting that Energy Tank came to be around 3 seconds slower but would allow to risk more in LN paying for the time lost.

Red Ki-Hunters rooms leaving LN: Behemoth had bad luck in these rooms which not only costed time but also a lot of health.

Kraid's Super Blocks Entry: If you are lucky enough to have a good number of Supers there you can shoot all three blocks and jump straight to the platform without rolling down like Behemoth did, but it's hard to be that lucky with drops leaving LN.

Power Wave set up in Draygon's fight: Health is very important at this stage to allow the reverse Full Halfie which allows to skip the Space Jump. By shinesparking Draygon when he is far of the wall you avoid getting a 40 health points damage by not having Draygon touching Samus hitbox. To have enough time for the super short charge under Draygon you have to deliver the second shinespark as soon as possible which ends up inflicting that 40 health points damage mentioned. The intention of the early Power Wave and shinespark is to allow Samus to recover faster from the crash animations so for the second shinespark you can wait till Draygon is far enough from the wall to shinespark, avoiding the additional damage while still beeing able to get in position for the third spark.

Big Metroid Skip: It can be done with, iirc, at least 2 jumps less. Behemoth had great runs (even better than this one) ending with a failed Big Metroid Skip. He decided then to go to safe mode there.

Considerations: If I tested it right a perfect run can be 1 min, at least, faster than the usual route. Zoast got a 44:08 but with more work he could definitely have reached sub 44. With that in mind we can say that sub 43 is possible, but it will probably be harder than sub 44 with the usual route.

I've seen some people saying Behemoth (and probably the ones that helped him with the run) kept secret from the comunity about his "discoveries" and stuffs like that. HOnestly WS first was ruled out for great speedrunners who didn't take the time to test the route properly by just assuming things. If anyone said "I'm going to try WS First" for any% he/she would probably just hear a "good luck with that" from others. There is a very old savestate run by Saturn posted on m2k2 that uses the very same route Behemoth used. Smokey presented this run though long after we had defined the best route, but it was after watching his run that we decided to look more closely into the Space Jump Skip. So the route has always been there to be used but no one gave a crap about it till someone got it and showed its potential. The idea was to deliver something new and refreshing to the comunity in a way that it would be pleasently surprised. Got this feeling many times before with TASes and speedruns (Hotarubi comes to mind) so we wanted to see this happen again and the best way was with a sub 44.

Hats off, Behemoth. Congratulations for this accomplishment and hope to see an improvement coming soon.

Edit: Just one more information: This run was actually the first sub 44 Behemoth got. Before that his record was 44:14. He got two 44:14 beeing the second one a potential 43:30/35 which got screwed by a failed Big Metroid Skip and a bad scape after Mother Brain.

Edit2: For those interested here is a link to a smv of the route in a theory tas route we did along with the emulator needed to play it.

https://m2k2.taigaforum.com/attachment/8i07zsOwIwDnELsAPQ3pzNzYtrM

Edit: Link fixed
Thanks Reeve for clearing those things up, you saved me a big post :P.

Just to extend a little on the X factor Draygon strat: Due to requiring at least 135/140 health for the return spark it's essential that you don't take too much damage in the fight, and as Reeve said to do this you need to activate the spark on Draygons head a bit later than usual so the crash animation doesn't sit on top of draygon and thus inflicting more damage. However the downside of sparking so late is that when you land after the crash, there is the danger that Draygon may swoop and if you begin short charging you'll run straight into draygon. At that point the space jump skip strategy is pretty much lost. That's where the beam combo-spark trick comes into play, to enable a quicker crash animation after the first initial spark on Draygon, so that when you fall and begin the next short charge, you will not get hit if Draygon swoops. On top of this the X factor manipulation trickery seen with the jumping toward the centre of the ring is actually much more consistent than you'd think, as long as you mimic the time in which I started charging the beam on the left and the rest of the movement cues afterward. It's quite common to get most, if not all particles to connect in this fashion.
Edit history:
zoast: 2014-08-30 09:42:39 pm
zoast: 2014-08-30 09:07:53 pm
Quote from Reeve:
I've seen some people saying Behemoth (and probably the ones that helped him with the run) kept secret from the comunity about his "discoveries" and stuffs like that. HOnestly WS first was ruled out for great speedrunners who didn't take the time to test the route properly by just assuming things. If anyone said "I'm going to try WS First" for any% he/she would probably just hear a "good luck with that" from others. There is a very old savestate run by Saturn posted on m2k2 that uses the very same route Behemoth used. Smokey presented this run though long after we had defined the best route, but it was after watching his run that we decided to look more closely into the Space Jump Skip. So the route has always been there to be used but no one gave a crap about it till someone got it and showed its potential. The idea was to deliver something new and refreshing to the comunity in a way that it would be pleasently surprised. Got this feeling many times before with TASes and speedruns (Hotarubi comes to mind) so we wanted to see this happen again and the best way was with a sub 44.


Some people at srl looked into WS first a year ago (or more) and it was written off.  I personally assumed more about this than it turns out I should have.  This is true. 

On the other hand, you guys did a lot of testing on this by your own account, not "hey I'm gonna try phantoon first".  And for me it depends on who says these things what kind of weight I give to it.  I mean, you did what you did and have done before, no need to rationalize it.

A lot of people have been pleasantly surprised by it but I have to be honest to you all, for me personally, the more I thought about it the more I got aggravated that you all knew for more than half a year that I've been grinding a suboptimal route.  And then I remembered behemoth even saying in my channel fairly recently that he wasn't interested in lowering the time for the game.  I know it's just a game and new possibilities are indeed fun, but damn guys.

Anyway, that aside, there is definitely a lot of sweet stuff in this run, and there's also a number of rooms that I don't agree with the strat choices.  One thing I will say I really think is cool is the phantoon strat, it will be a nice addition to low% in the future.
Edit history:
Reeve: 2014-08-31 05:54:40 am
Reeve: 2014-08-31 05:34:04 am
Reeve: 2014-08-31 05:32:29 am
Reeve: 2014-08-31 05:29:42 am
Reeve: 2014-08-31 05:20:34 am
Reeve: 2014-08-31 04:16:04 am
Reeve: 2014-08-31 04:14:51 am
Reeve: 2014-08-31 04:14:10 am
Quote from zoast:
A lot of people have been pleasantly surprised by it but I have to be honest to you all, for me personally, the more I thought about it the more I got aggravated that you all knew for more than half a year that I've been grinding a suboptimal route.  And then I remembered behemoth even saying in my channel fairly recently that he wasn't interested in lowering the time for the game.  I know it's just a game and new possibilities are indeed fun, but damn guys.

I'm not talking for anyone here except me, I do understand your reasons to be upset, but in the other hand what you did to this game is not just lost because a new route was found. Some time ago no one would ever believe it would be possible to reach 28min in-game time in an in-game time oriented run, let alone a run that is highly optimized for realtime without any pause screens. You don't even have to say you don't give a shit about in-game time, we all know, but what you did is valid and the Super Metroid comunity will keep talking about it forever. The first man to reach 28min in-game time and it will be probably only you for a long time (hopefully not forever). Anyone who wants to beat you will have to do it in the standards you set, so with a 28min in-game time with a realtime below 44:08. Honestly I think you should submit your run to SDA, but that's your choice. I honestly respect you a lot and love what you did to my favorite game of all times.

Quote from zoast:
Anyway, that aside, there is definitely a lot of sweet stuff in this run, and there's also a number of rooms that I don't agree with the strat choices.  One thing I will say I really think is cool is the phantoon strat, it will be a nice addition to low% in the future.

Really cool you liked the run and indeed there's lots of cool stuffs. Shame behemoth didn't nail some other strategies he intended to but there's a smv I posted right above with most of the intended tricks. There's also some behemoth decided not to use for personal reasons (consistency most probably) but, again, you are a fantastic player along with many others in the SM comunity and what I most want is to see the new levels of optimizations this route can reach. In regards of Low% yeah, the momment Behemoth came up with the Phantoon strategy I thought, "oh shit, this is gonna save a good chunk of time for Low%". Funny this strategy shooting the first flames right when Phantoon open his eyes is not new. I've seen it before at least once by a runner in a savestated run. I honestly assumed the guy optimized the run too much and it wasn't something that could be consistent. We all know now that we shouldn't assume things without testing.

I'm posting a small timesaver I came up with for Hi-Jump room and a smv of a safe way to damage boost from the spikes on the ceiling of the spiked room in Red Brinstar after WS.

AfterHi-Jumps: The smv explains itself.

RBSafeDamageBoost: By pseudo screw attacking Samus is not damaged by the spikes on the ceiling. If you charge your beam and jump to the spikes, you can correct Samus position before taking the damage. After correcting Samus position, while still holding jump, press up or down for samus to get the damage and quickly press backwards. You can keep holding the beam till she starts flying backwards so she kills the crab after landing. Sometimes Samus land in a position that, when she starts running, she get's damaged again by that crab.

Another safe way of doing this room is to not damage boost there. Instead charge your beam, jump as late as possible and as high as possible, then release the shot button so Samus will shoot as soon as she touches the ground. She will then kill the next crab on the ground and you can just gap skip to go ahead. This will also allow you to get a drop on the way to the door.

Use the same emulator I posted in my previous post along with the savestated run.
Just to echo what Reeve said. In game time is still absolutely relevant and evidence of this was your continued effort for the 28 which was fantastic work. We knew you had 28 in the bag and you more than anyone was going to achieve it, it needed to be done. This new route is kind of like the answer to the question, "so what's next for any% RTA?". The timing of it was rather iffy i know, just after you got the 28 this new route shows up. but you have to understand I had been trying to get a 43 for ages with this route. If I got it much sooner I would have presented it then and there before more progress had been made by you and Kott on the current route. Once again as stated, the goal was to surprise the community harking back to the old days and the best way was with it also being a 43. I understand being upset, but honestly this was well intentioned.

About the run itself. This run was tested so extensively. Things can look a bit dubious at times like shooting the pinks doors with missiles in Norfair instead of a super. Starting from Gravity suit you only lose about 3 seconds for all the farming on the cacti's in Red brinstar and the pink doors with missiles in Norfair. Saving you about 4 supers is great way to only have to spend a few waves at the bugs near speed booster instead of getting robbed standing there for over 6 rounds trying to get to 10 supers from a measly 1 - 2.

traversing LN on the way to Ridley. More risk appears to be potential improvements here but it's really asking for trouble to do so. With the ki hunters outside save room and the giant hoppers at the kago room, you just have to wait for them to get out the way before moving as their positions are too much of a roadblock and taking damage would be a critical failure at this stage.

Bombing instead of power bombing blocks escaping LN. The reason for the 3 blocks after kago is because the PB generates so much lag that it only saves about 10 frames. Also bombing provides better positions for the hunters so you can skip past the first one and sometimes the third one further up, preparing you for the Maridia sections later. I did mess up and took too much damage here, but the idea is to only take 1 or 2 hits leaving LN, then Kraid/Maridia is all the better for it afterwards. This is why I had to farm slightly after Botwoon.

Bombing pillars instead of a single PB when re-entering upper Norfair. When I tested it I was surprised to see that it was slower by around 20 frames to use a PB due to the lag by the time you leave the tunnel when mock balling. I guess it's possible to get that orb boost onto the platform if you use a PB though, but meh.

Tourian strats: This route nearly always makes you enter Tourian with very low supers due to skipping Draygons drops and other things, so my strats I had trained myself to do for so long was pure gathering strats on the metroids in all the rooms. In this run however due to decent RNG toward the end and extra farming I had loads of supers, so the strats looked out of place with the amount of ammo I had. Still, they are not much slower due to having no space jump anyway, and they really help not being robbed on super drops.
Edit history:
Smokey_SDA: 2014-09-01 10:56:30 am
Smokey_SDA: 2014-08-31 07:55:27 pm
Smokey_SDA: 2014-08-31 12:12:46 pm
Quote from zoast:

Some people at srl looked into WS first a year ago (or more) and it was written off.  I personally assumed more about this than it turns out I should have.  This is true. 

On the other hand, you guys did a lot of testing on this by your own account, not "hey I'm gonna try phantoon first".  And for me it depends on who says these things what kind of weight I give to it.  I mean, you did what you did and have done before, no need to rationalize it.

A lot of people have been pleasantly surprised by it but I have to be honest to you all, for me personally, the more I thought about it the more I got aggravated that you all knew for more than half a year that I've been grinding a suboptimal route.  And then I remembered behemoth even saying in my channel fairly recently that he wasn't interested in lowering the time for the game.  I know it's just a game and new possibilities are indeed fun, but damn guys.

Anyway, that aside, there is definitely a lot of sweet stuff in this run, and there's also a number of rooms that I don't agree with the strat choices.  One thing I will say I really think is cool is the phantoon strat, it will be a nice addition to low% in the future.


My post will probably look redundant now but I wanted post this so you and others can read it from me. You shouldn't feel aggravated by this at all. You did a tremendous work on the :28 game time and I personally followed your streams heavily this last month, excited to see the :28 that was very close for you. Nothing is ever lost when grinding this game and the more people push it to the limit the better because then standards are set. As for the WS-first route, the route has been talked about for more than 10 years and no one (apparently insane enough) has wanted to capitalize on its potential for console conditions. Before we did any major testing, I think Behemoth openly expressed to the SM community his interest in the route and that he was working with Reeve and me over at m2k2, and no one showed the same interest to further join in the discussion and research. So we kind of took matter into our own hands, so to speak and it was a lot of work for the three of us to make the route playable for console. The private thread was started by me telling Behemoth about a brief, fun return I wanted to do, and later developed into the WS-first route discussion. The forum (it doesn't matter if it was m2k2 or not) allowed us to keep track and document the research of the route in a much better way than a chat based discussion.

After we realized the potential of the route, we wanted instead to surprise the SM community a la hotarubi or TAS run as Reeve and Behemoth have said, to show its potential and to show how limitless the game still is for console. Without a console run and words alone, people would not believe us. So you have to understand our reasoning for why we kept this private until the 43 emerged. But the timing was pretty bad with your :28 and Behemoth's 43, I understand that, but please understand that Behemoth was grinding this since earlier this year. I'm pretty sure that if you go for it you'll master it very soon and I hope to see some runs from you in the future. As for giving weight to what others have to say about a route or anything else related to the game, it's a very understandable mistake we all make.

Hope I didn't sound like a prick by any of this, it's not my intention. Apparently the "m2k2" community are us three, and we have nothing against SRL. Congrats on the :28! I'm glad and it's cool that you liked the run/route and that you'll find new strats for it, and I hope you grind this route as well, as it's real time oriented.
behemoth - holy sh*t again!!
Edit history:
Smokey: 2014-12-28 10:24:13 pm
Smokey: 2014-12-28 10:21:10 pm
Smokey: 2014-12-28 10:19:47 pm
Smokey: 2014-10-11 04:19:06 pm
Hello.

Felt like posting this here so it might be useful to anyone interested. Attached is an emulator run like the ones I used to do a little while ago, which involves using savestates during door transitions 99% of time, to do each room as clean as possible like any top player would at normal execution, and other savestates during boss fights (not to take advantage of patterns) or past certain events merely for commodity reasons and to speed up the process of doing the run.

http://www.twitch.tv/smokeysda/c/5769014

It uses the "new" RTA route which we use to call it WS-LN for short, with a finishing time of 41:29 RTA by SRL rules (ntsc). This is by no means a theory TAS run like most folks are familiar with, since the purpose is not to save every possible tiny frame in every room, but to represent generally the potential that the route can offer at its best of normal play. For those interested the run shows a few recent little strats that Behemoth, Reeve and me have incorporated, but we still consider this a WIP as we think the route is still in its infancy so new developments are to be expected (as always) by the community which can improve this run or the route in general.

As for the run itself (people might already know most things):

- It is important to conserve health after the first ET since the route leads directly to the WS after Brinstar.

- I don't have anything to say about CWJ as the only way to get it to work is by practicing the hell out it and developing your own cues for it. One thing to remind is the option of DBJ. If you're good at BJ this is it, and I rather find this method more consistent than CWJ as I think there are a few more frames you can juggle with for the bombs placements as long as the rhythm is correct and if you quickly spot an error on the BJ you can land quickly on the pillar again to retry. I think this is the killer of the run for me. LN over time can become adaptable with practice and decent play.

-  Phantoon: For the eye-stab, it's important to learn Samus placements in respect to Phantoon's attack when shooting the circle of blue flames. Whenever he shoots the regular flames shoot 'em diligently, you'll want the room as clear as possible for the eye-stab and the missile attacks. In this particular case I had mid/mid/mid which was excellent for health/ammo drops just before shooting this dude. If you get to the fight with good health/ammo, a decent fight would be fast/mid/fast or fast/mid/mid but that's if you're lucky.

- Attic: this place is expected to have developments over time, especially at the far left section of the room when killing the remaining enemies.

- From red-Brinstar to Wave Beam: it's important to conserve supers ammo for Ridley. Notice that most rooms with pink doors are opened with missiles and it's best to leave the pipe bugs farm after Sbooster after 3 waves (they have a nice habit of not dropping supers so this is why it's important to conserve them). Anything more than 3 waves, it's time lost. The best scenario is having 9 supers for the fight after 42 charged shots. For every super you have less than 9 for the fight, you have to add 2 charged shots after those 42 shots before using supers.

- It's debatable whether it's faster to grab the Ice Beam before or after getting HJB, since the same distance is traveled. Behemoth did some test on this before, and getting HJB later is faster by a few frames, possibly less than a second, and it's probably a bit faster from the way you can climb up the room after HJB to get directly to the first big heat room. 

- HJB room exit: self explanatory. I find it's more easier to do this by delaying a bit the back-down fall.

- You have to monitor your PB ammo as well, as you want to enter LN with at least 3 of them. Good spots for PB ammo on-the-fly are killing the walking Dragons (seahorse like enemies before Wave) and of course the blue orbs at the yellow door before LN.

- Worst room in the game (vertical LN room with pirates): afaik, using the missile at the top reduces the lag when opening the door (probably a bad decision was using the item-cancel for the missile in this run tho).

- The vertical room with the pirates by the save room: a nice addition by Reeve, it's about a second faster than waiting for the pirate to roll under it. It's somewhat risky at this stage but definitely worth it if your health is decent. Visual cue is easy to spot, and it's consistent as long as you get down there in time.

- Skipping Ridley's ET (replaced by Botwoon's ET): you avoid 2 door transitions and it's faster by about 3 seconds with perfect play all the way. Behemoth did this in his earlier runs only if Ridley left good health drops (replenishing the 2 ET's) but this meant careful play ahead which was slower and less reckless when escaping LN. And since we can't be certain of the obstacles and RNG ahead, getting Ridley's ET and skipping Bot's ET just averages the same as not doing that door transition and magically escape LN with decent speed and health.

- Reverse kago-room: this dboost is a bit complicated to understand but the trick to manipulate slightly the hopper's RNG is to ledge grab the top platform before the PB explodes, as this allows the hopper to move in such a way to facilitate the dboost.

- Reverse kago-room exit door to pirates room: by entering the door in morph ball form it allows for the first flying pirate in the next room to get out of the way. By doing this you avoid a hit, you can even avoid the hit of the top pirate if you're lucky (the intermediate one isn't a huge problem almost all the time).

- Reverse mini-Kraid room: possibly one of the hardest rooms to get right, a nice way I found to setup the dash for the morph ball in next room is to charge the beam by the time you enter the door. The pirates' placements in this room is pretty annoying and this eliminates all the uncertainty.

- Glass tube: pretty much another place that can be polished, as there's no consensus as to what's the best possible way to reach the upper door.

- Draygon: as noted before, just make sure you get out of there with 140 of health or more to skip the SJ and successfully do shinespark on your way out. Initially, skipping SJ saves about 20 secs by avoid the item box and 2 door transitions, and with decent play (until the end of the game) you lose almost as 8-10 secs from those 20 secs by not having it. So in average, you gain about 11-12 secs by skipping it.

- Metroid skip: for the airball, the setup for the initial jump was to wait until the metroid reached the highest floating point (or as soon as the metroid starts moving the claws) to start the dash for the airball. By doing this you don't have to jump too late or right before the pipe on the ground. You can do the jump a very good frames earlier and this allows for more relaxed timing and play. Next, do the airball before touching the ceiling and then land in crouch position. Overall, the skip can be done in 4 or 5 jumps and with no lag from the dead enemies, as long as you don't touch 'em of course.

Anyway that's all I can think of atm, hope it helps.
It doesn't look like this thread is super active anymore, but I can't think of a better place to try and start discussing speed runs for this amazing game than right here! So here goes. My brief history with this game:

-First played at a time when I was too young to remember my specific age (somewhere between 6-8 years old probably.) Fell in love immediately. I am 26 now.
-One of the first games I remember buying with my own money to own for myself
-First saw a speed run of the game around the year 2003-2004, right here on SDA. Became one of my favorite videos ever. I was so impressed I started trying to imitate the runs.
-Got my time down to a respectable ~1:18-1:20 time without doing any actual research on speed run methods/tactics/etc (there was a lot less of a presence online for this sort of stuff too if I remember correctly)
-Stopped playing as much consistently due to life and other happenings.

A couple years back (maybe it was just a year ago?) I saw the 4 way race on AGDQ and realized just how out of touch I had become with speed running this game. People were doing things like the full halfy, the blue suit glitch on Draygon, etc etc. I once again was mesmerized by what I was seeing and watched that race video probably 5 times in 3 days. Life and other happenings kept me from playing the game much, but I did actually boot it up for the first time in a few years and, using my limited memory of how I used to run the game, made it about halfway in.... 50 IGT minutes. Ugh. So rusty.

This past Christmas break I decided I was going to start speed running the game again. I watched the 100% world record video on SDA for research and began trying to imitate the video. It became clear to me quickly that I needed more than just visuals to figure out how to do a lot of this stuff. So I looked online and found out about stutter stepping for super short shinespark, Phantoon tactics, other boss tactics, exactly how to glitch Kraid so he didn't crash through the ceiling, general movement tactics, how to consistently damage boost, etc etc. I've run the game 6 times in the last 2-3 weeks and I'm going to start tracking realtime soon, but I've still been abusing save states to the point I'm not quite comfortable with that yet. My PB (using save states..) so far is 1:04. I can feel the sub 1:00 IGT within my reach. Something I was initially shooting for back in the day and eventually gave up on. But I'm curious what the community has to say about what I might try to shoot for this.

I feel like I've got most of the "tricks" down, with the exception of the full halfy which is pure insanity. I can't quite manage to scale the wall and get through the door fast enough to use a shinespark, let alone jumping into position after entering the second room to pull it off. But I can:
-Consistently kill Kraid correctly. (no crashing through the ceiling)
-I've got stutter stepping down for things like Maridia Mainstreet, Blue suit glitch on Draygon, and the Crocomire exit (to get power bombs).
-Phantoon is one of my rough spots, but I feel like that can be said for everybody lol. And I've had "perfect" runs with him.
-I can do the Metroid skip (not 100% of the time, but I've made it to the door a few times).
-I can also do the Zebetite skip (not without disabling Screw Attack though... can't quite do the stutter step down there for some reason and I wasn't even aware you could do it without disabling screw attack via other methods until AGDQ 2015 the other day. Still not exactly sure how sweetnumb did it.)
-I had ocean fly down, but apparently people don't use that anymore? Sad day... it was so fun.
-And it goes without saying I can do things like the machball for early super missiles etc.

I've probably missed some stuff, but I feel like that at least confirms that I *kindof* know what I'm doing haha. So when I'm trying to think of other ways I can reduce my time, I guess it all comes down to basic speed tactics... general movement, damage boosting more often/better, etc. I find it hard to believe that there's a discrepancy of 15+ IGT minutes between my runs and the WR videos I've seen. But my most recent run I literally shadowed the video for 90%+ of the rooms using save states and I still had a time of 1:04. Obviously it wasn't a perfect recreation of the run, but I tried to use the same routes and tactics for every room and that's the time I got. So.... general movement is the only thing I can think of to blame. Sad

I'm not giving up. I love the game way too much to stop when I'm so close to a sub hour run. It is my favorite game of all time (tied with Metroid Prime) as well so it's not like I'm gonna burn myself out on it. I just have to keep finding time to play and convincing myself I'd rather do that than any of the other stuff I could be doing instead. Smiley

If it would help, I can record a run or a section of a run for feedback. I'll be surprised if anybody comments back today or anytime super soon but I figured I'd reach out anyway. Also, I have to say I read through the entire thread and I love the general feeling of community you guys exhibit. I'm proud to associate myself with a group of such passionate people about the best game ever made. Smiley anyway I've droned on for long enough. Thanks for reading and let me know if you guys have any tips or "ah ha" moments that maybe helped you with your own runs!
Edit history:
Antilles58: 2015-01-13 09:34:53 am
#FailFish
Wow! You can do a lot of stuff, and have learned pretty quickly!  Well done!  What you've done is put together a pretty decent segmented run (ie, the going back to save states and just taking your best segments) - but it's time to move on to single-segment RTA! Smiley

Using save states is a great way to start, but eventually you want to move away from that since executing things in a real run and executing them out of a save are very different from a difficulty perspective.  Half the battle with speedruns is recovering from your mistakes.

If you looked online, you probably found this Wiki: http://deanyd.net/sm/index.php?title=Main_Page - it's the GO TO place for tips.

I highly recommend that you find Go1den's tutorial on Twitch and watch it - even if you already know those tricks, it's a great to review.  Even though it's any%, all the stuff applies to 100% too - you just route it differently and have a little more wiggle room.

Once you've taken that, I recommend watching Zoast's tutorial - this thing is an 8-hour marathon, but it is literally a Master's course in Super Meroid speed running.

I also recommend Zoast's 100% run from SGDQ 2013.  That's a very clean run, and some of the strats used there aren't as complicated as what sweetnumb was pulling off this year.  But both GDQ runs are good because the commentary helps give tips.

Lastly, get over to SRL and join the weekly Friday night any% race!  It's a ton of fun - I almost always fail and have to forfeit, but those times when I succeed are oh-so-good!  Oftentimes there is a 100% race right afterwards.  Talking to people in the SRL #supermetroid channel and before the races are also good ways to get tips.
#FailFish
Some other advice I got was that boss battles trump specific room strategies.  Those are probably the biggest places you can save/lose large chunks of time, whereas any individual room isn't probably going to cost you that much.  Your overall movement will improve as you keep playing, and you can save minutes off of your time by getting quicker boss fights.
Wow! Thanks for the quick response! I love the deanyd.net site, and have definitely already used it as a major resource for learning this stuff. I wasn't aware of the tutorials you mentioned, so I'll be searching for those momentarily. Is Zoast's tutorial just on twitch like Go1den's? I'm pretty sure Zoast was the one who did the video I fell in love with back in the day and I love his more recent runs, so I have a bit of a soft spot for his work (as it appears many people do lol).

I'll definitely check out the SGDQ 2013 video as well. And as far as the weekly Friday night any%, that sounds like an awesome weekly tradition and will hopefully spruce up my Friday nights haha. Thanks for all the resources and tips. I wish I wasn't at work so I could get to "work" on all this important stuff right now haha
#FailFish
Oh man I know exactly how you feel.  My main game is Donkey 2, and I've got a couple PBs that are about to fall - all i want to do is get there and knock those suckers down!  My wife was on board with the idea of a sick day to game, but I don't think my employers would be.... *sigh* at least there's the forums!

Yes, Zoast's tutorial is on Twitch as well.  I can't access it from work or I'd send you a link to both of them (I'll probably post that later when I get home).  Zoast literally goes through every single room step by step by step.  Again, it's any%, but the general movement is still relevant.  I downloaded the videos and burned them to a disk so I could watch them here when I had down time.

The weekly race is every Friday night at 11PM EST.  SRL isn't too hard to figure out, but if you need more help, I'd be happy to offer it.  I'm pretty new, myself - only been here a little less than a year, and I'm still awful at SM - but I'm happy to help where I can!
So I started a new run last night with the intention of using no save states. I did relatively well up until Phantoon which of course I totally botched. Then I died in the spike platform room right after him. ugh. So I started another run immediately and.....same thing happened haha.

I took a break and then did one more run, managed to get past that and grab the Gravity suit, and get back to my ship (which was my first checkpoint in my other runs). I recharged and saved because I was tired and wanted to see how I was fairing, and compared to my save state runs I only lost 4 IGT minutes. That's obviously more of a discrepancy in real time but still not bad I guess. I'm currently struggling to do the criteria beam wrap glitch with just the wave beam. I can do it with spazer no problem, but I don't have a consistent method down for wave only... and I'd rather avoid adding another pause to the run if I can help it since I'm gonna start tracking real time. >.<

Anyway when I finish the run I'll probably update here about how things are going etc etc. Thanks again for your links. Go1den's tutorial is making me feel better about myself (lol). I haven't been able to watch Zoast's yet.
#FailFish
Nice!  That spike room is the last major run killer in 100%.  Once you get past there, you're pretty much home free.  My last 100% died in that room as well - it's tricky to do well.

But you should always have spazer at that point in 100% - you should get it on the way to Kraid.  Hopefully that helps.

Speaking of wraparound glitch - where did you find a good tutorial on that?  I haven't been able to find a clear guide anywhere.  It's probably one of the biggest tricks I'm missing that's costing me a lot of time in that run (general bad movement aside).
I have the spazer, but it's disabled for the Phantoon fight, and I would prefer not to pause again to re-enable it if I can avoid it.

As far as tutorials for the wrap around glitch goes, I just copied one of the runs I saw. Might have been Zoast. They did it with Spazer whoever they were, I just stood in the spot they did, and released shoot and run at the same time. It took probably 10 minutes of practice and then I found the "sweet spot"
#FailFish
Guess I just need to grind it out then. AGH! TOO MANY GAMES TO PLAY!  At least I got a new 102 PB yesterday. Wink

I'm with you on not wanting to pause for RTA (but if you're going for in game time, pause doesn't affect that).  What did Zoast do in his run - did he pause to re-enable spazer somewhere after the ghost?  I'm pretty sure I remember that phantoon fight using the x-factor method... now I'm curious.  I almost wonder if spazer or plasma is required for that trick.  I seem to remember seeing it in a video somewhere.
Edit history:
Antilles58: 2015-01-14 10:06:13 am
Antilles58: 2015-01-14 10:03:51 am
#FailFish
Here's two videos:




They're both pretty old, but the second one by dessyreqt seems to definitely indicate plasma or spazer are required (in the video description).
#FailFish
Speaking of videos, here's another (old) tutorial by Dessyreqt.  Lots is pretty different now at the top tier, but for starting out, this is pretty darn good.

Yeah the 2 wrap videos you posted I found last night. I figured you had to use spazer until sweetnumb's agdq 2015 run. He does it with wave, and I did get it *once* last night when I was trying for it. It just seems much harder. I'll definitely check out the 100% tutorial as well. Thank you!
#FailFish
Oh dang I didn't notice that sweetnumb did it with just wave!  That run was so hype - sub-1:20 WITH the animal save. 

Poor, poor Epic.  Most heartbreaking moment ever.
I audibly gasped when I saw it. I felt so bad for him.... Cry

Anyway yeah. I'll probably put another 10-15 minutes into trying to figure out wave tonight and if I can't then I'll just eat the 5 seconds or whatever it takes to switch. I prefer Spazer anyway, the problem will be switching back again for the Maridia fights.
#FailFish
It's good to practice real strats, obviously.  But look at it this way - 5 seconds in the menu isn't really that big a hit when you have minutes to save elsewhere on the run.  Work on adding strats that save seconds when there aren't minutes to make up elsewhere.  That's my feeling, anyway.
Any chance you could link the Zoast tutorial you mentioned? I spent awhile looking for it and haven't been able to find it. Taking my lunch break now. Nintendo needs to release this game on 3DS so I can play it on the go... >.>