Hey, so I was really into grinding individual levels of this back in 2012. Any% never really piqued my interest in this game so I always used to just grind levels for 5 stars as fast as possible. I recently dabbled with the game again but quickly lost interest in it as a speedrun, much like I did 6 years ago. However, I have acquired some knowledge of the game over the years that I figured I would pass on. Do with it what you will.

First, I did roughly 50 attempts of Born Again (5 star) runs the other night and managed a 4:46 (link posted below). I have no interest in posting it to the leaderboard and I have absolutely zero interest in doing full-game runs or grinding other levels. Again, I just get bored of this game pretty quickly these days. I'm actually hoping that someone can take my strategies or whatever and get it down to sub 4:40, as I had a run with better luck (will explain in a moment) where I screwed up the hand-to-hand fighting portion. That would have gotten me under 4:40, so I'm hoping someone else can do so.

The run can be found here: https://www.smashcast.tv/HavocProdigy/videos/1391197

So for Born Again, really 7 of the 8 hunters can be killed the same way each time, without any degree of randomness. The only problematic one is the 2nd one that I kill in the van section. In the botched run I had that should have been sub 4:40, I got the "god pattern" where he walks right, going right past the gate where this section starts. He can be killed with a low-level execution immediately if this happens. However, he rarely walks that way.

Secondly, being unable to crack the style points rating system was a big deterrent for me back in the day. There are many stages where you can get 5 stars without a single execution. However, the ideal execution amounts for Born Again is 4 reds, 2 yellows, and 1 white. You can also do 3 reds and 4 yellows, but from my testing, this is slower. There is zero reason to do 5 red executions on Born Again. It's just wasted time.

Lastly, I can provide the kills versus execution numbers that I used to use on all 20 stages to yield 5 stars. Of course, this won't tell you how many red, yellow, and white executions you need to do on each stage. I was able to crack the code and find the minimalist amount for a few of the stages, but I lost interest well before I could figure them all out. Of course, none of this really matters for Any% IL grinding or Any% runs, but if anyone wants to use this information for 5-star ILs or runs, it would probably save you hours and hours of testing. Enjoy.

Born Again: 8 (7 executed and 1 killed) Doorway to Hell: 10 (8 executed and 2 killed) Road to Ruin: 11 (9 executed and 2 killed) White Trash: 16 (12 executed and 4 killed) * Fuelled by Hate: 24 (16 executed and 8 killed) * Grounds for Assault: 17 (15 executed and 2 killed) Strapped for Cash: 19 (13 executed and 6 killed) View of Innocence: 26 (26 killed) Drunk Driving: 19 (13 executed and 6 killed) Graveyard Shift: 35 (35 killed) Mouth of Madness: 25 (17 killed and 8 executed) Doing Time: 21 (13 killed and 8 executed) * Kill the Rabbit: 49 (49 killed) Divided they Fall: 27 (23 killed and 4 executed) * Press Coverage: 32 (27 killed and 5 executed) Wrong Side of the Tracks: 25 (25 killed) * Trained to Kill: 28 (25 killed and 3 executed) Border Patrol: 27 (23 killed and 4 executed) * Key Personnel: 32 (27 killed and 5 executed) * Deliverance: 7 (4 killed and 3 executed) *

-White Trash: has 14 enemies if you don't pick up the nail gun in the trailer. By picking it up, 2 enemies spawn outside, bringing the total to 16.

-Fuelled by Trash: Your results will vary depending on whether or not you stay in both cranes to kill the hunters. I do not kill enemies with either crane. By exiting the first crane, you will only fight 2 hunters instead of 4. By staying in the second crane, you will fight 5 enemies (or 6 but I'm pretty sure it's 5.) By exiting the crane, you will encounter anywhere from 3-5 hunters, seemingly depending on where you're standing. I learned in Border Patrol that the respawning enemies don't seem to respawn if you're standing by their typical respawn location. Sometimes I encounter 3 hunters when I exit the crane, sometimes 4, and sometimes 5. After this part, there's still the hunter with the nail gun that guards the exit. That part does not change. So on a given run, I will encounter either 24, 25, or 26 enemies. This is the only level that I cannot seem to completely control the exact amount of enemies that I will face.

-Doing Time: I'm sure everyone is aware that the enemies continue to respawn until you meet the violent and gruesome execution requirements at the beginning of the scene.

-Divided they Fall: After killing Ramirez, I'm pretty certain there are 3 enemies remaining that I don't kill. I don't know the precise amount of enemies in this level but I believe that it is a finite amount.

-Wrong Side of the Tracks: There are actually 28 enemies in this level but after turning the power back on, I slowly creep towards the starting station as 3 enemies enter and sneak onto the train unnoticed. I don't bother killing these 3 because it is not worth it.

-Border Patrol: I used to kill the enemy by the stairs and then kill the two snipers (which would be kills 18 and 19) and then kill all of the remaining enemies. This brought my total to 25 which apparently is not enough for 5 stars. This level has some kind of issue. I can kill 26 enemies each and every time I play it, but the game will tell me I killed 27. So anyway, I don't remember all of the details here, but I had to let one of the enemies near the mansion respawn in order to get enough kills for 5 stars.

-Key Personnel: The enemy amount is not infinite but results will vary depending on how many cameras you are spotted by. In a typical run, I shoot out only the camera nearest to the lead Cerberus and am never spotted again. However, in a test run, I was purposely spotted over and over and ended up with around 60 enemies before they stopped respawning.

-Deliverance: Piggsy and Starkweather both count as executions so it is impossible to complete the level without at least 2 executions. This seems to be enough to get all of the style points so I don't believe it is possible to get less than 3 stars on this scene, though I cannot say this with 100% confidence.

I still hold the unpopular opinion that this is a really solid NES game, although I understand some of the criticisms and don't really disagree with a lot of them. I have to laugh when people's number one complaint is that the game is unfairly cryptic, as if Zelda 1 and 2, Metroid, Simon's Quest, etc. didn't have the same issues. Those games were/are just more popular and so information among your friends and whatnot was much easier to come by back in those days for such games. But I digress...

My only real motivation for grinding this again is that I don't have a recording of my best time. On a whim, I just decided to do some attempts back on April 10th after a lengthy break from the game (and from speedrunning in general). I don't know if time away just helps to reset one's mindset or if I just hit the lottery on that particular attempt. Since my 9:03 in May of 2014, I've put in a few thousand attempts and for the most part, never came close. With the WR route and strats, I've completed MAYBE 10 runs since then. Prior to that, my two best times were both in May of 2014: the 9:03 and a 9:25 about two weeks before that. Anything under 10 minutes is pretty god-damn difficult and it's certainly not because there's not time to save (there's huge chunks everywhere). Surviving this run with an aggressive route is...problematic.

I didn't record the 8:50 because by that point, I hadn't really done any speedrunning since October of last year and it was just sort of a "let's do a few attempts for fun" sort of thing. Hell, it was on emulator anyway. I didn't really feel like hooking up my recording stuff. There was really no reason for me to believe that I would PB or even complete a run. The pieces just fell into place.

Anyway, the one good thing about this is that there are several newer strats that I use that aren't found in the 9:03 run that I would love to implement into my final PB. There's now several chunks of significant time to save in the middle of the run. However, the beginning and the end of the run are pretty difficult to compete against. I'm pretty much willing to settle for any minor improvement that's recorded on console.

I'm still considering uploading T_fumanu's run. I've seen a run or two of his posted elsewhere on this site by others, although I'd feel a lot better about it if I knew he was OK with doing so.

Thanks for the interest.

It's honestly not an ideal speed-game, as it's unbelievably difficult to complete with an ideal route. That, in conjunction with the negative reputation of this game, have ensured that it will never be a popular speedrun choice.

When I started speedrunning this game in 2013, I found a leaderboard (now defunct) that I had to translate from Japanese to English. At the time, T_fumanu's 9:54 RTA run (equivalent to 9:46 SDA timing) from June of 2012 was the best time listed, although no video was available. The runner finally made the video publically available around the time I got my 9:03. His run can be found here:

The other times listed on it were a 9:57, 11:10, and 13:01. I still have a screenshot of the leaderboard, although the site itself no longer exists. They also have a category for an 8-loop run (kind of like what happens in successive playthroughs of Ghosts 'n' Goblins) where T_Fumanu has the record at a ridiculous 4:54:07. I happened to watch the end of that live on Nico back in 2014.

I would love to add other people's runs, particularly T_fumanu's run, as it set the bar incredibly high and he has recorded proof of his run. However, without their permission, it's kind of a slippery slope. So yeah, even though there have been some others that have the run game, without their voluntary submissions, they're probably never going to be posted here. If I did the same for the Little Nemo leaderboard, there'd be at least 10 people off of the top of my head that I could add to the board that have runs publicly posted elsewhere.

Basically, everyone from the Western World that attempted to learn to speedrun this game made the intelligent decision that it's not a worthwhile endeavor...except for me.

P.S. I got an 8:50 back on April 10th but regretfully, I don't have a recording. I'm going to try and beat that run sometime this year. I don't really do much speed-gaming these days but I think I still have a PB left in me for this game.

I think we're in agreement that the run is clearly submitted in the wrong category and that 2-Boss Up+A/Deathwarps/Whatever should be a new category so hopefully that will put the issue to rest when a capable mod comes along.

Discussion on the rules for current/future runners would be great. Breaking the rule because you don't like it and then trying to justify it afterwards really isn't cool, regardless of who is going to come in and try and justify it on your behalf. I don't know you well enough that I'd bother to hold it against you. I think it was a poor choice and I wasn't going to remain silent about it but I think everyone has the same intended solution to the issue and so I don't have anything further to add. I hope it works out. Peace.

You don't get to decide that the rules for an established category don't exist anymore because you don't like them. If you want to have a discussion about it, then cool, discuss it. Submitting a run which blatantly disregards one of the only two stipulations of the 2-boss run (Beat Kraid + Ridley with no deathwarps) and proudly declaring you have "world record" is about the scummiest approach you could take. The fact that a mod that clearly doesn't know the run at all approved it on the spot doesn't make it acceptable.

There's a deathwarp and a no-deathwarp variation for both any% and 100% and so if there are people actually interested in running 2-boss with deathwarps, making a deathwarp variation for 2-boss seems like the logical solution. Deathwarps have always been considered a slower version of Up+A...they establish the exact same outcome. I would urge one of the mods (hopefully someone that's pretty familiar with Metroid NES speedruns) to make a 2-boss category that allows Up+A/deathwarps/whatever you want to call it. I find it hard to believe that there would be any objections to that solution.

As a further example, Inkject's best run of Metroid was while attempting Any% No Up+A. Unfortunately, somewhere during the run, he dies (it's been a while since I've seen it, I believe it was on the way out of Kraid's lair). He ended up PB'ing but correctly submitted it to the Any% Up+A category since he deathwarped in the run (unintentionally). What he didn't do was submit it as an Any% No Up+A run...because it wasn't. Deathwarping and Up+A do the same thing.

Tell that to everyone that has run Any% no up+A. You basically just decided that that entire category is archaic as well and that anyone who feels like it can submit a run with deathwarps to that category and obsolete better runs. It's like these people think that they just discovered deathwarps in this game and that magically, no one else ever thought to do it. Others weren't doing it for a reason and it's not because they didn't think to do so. You don't get to change the rules on a whim and then tell others why it's ok that you and you alone determined that the rules for the category shouldn't apply anymore. Way to tell runners like Cak, Controllerhead, and Dookel, who have put hundrends (probably thousands) of hours into any% (no up+A) and 2-Boss (no up+A) that their runs don't matter and that they were doing it wrong.

"It's faster" doesn't always fly in categories that aren't any%. Sorry. If a lower time is all that matters, resubmit your runs to any% up+A since that's the only category that they should have been placed under. Try submitting a 70 Star run of Super Mario 64 with backward long-jumps and Mips-Clip and you'd be rejected in a heart-beat and laughed at for your ignorance. The same thing should have happened here.

The fact that Big 20 runners are coming into the established Metroid NES community and determining that the rules for their silly race should take precedence over the long-standing rules "because it's faster" lol is disgusting. Get over yourselves.

Why are runs with deathwarps being approved in this category? It's insulting to the people who have actually spent a lot of time on the run and have done it the correct way. Deathwarps have always been banned in 2-boss.

I'm so sorry that my 2-sentence-long post inconvenienced your busy life that horribly. Searching for an answer brought up various "broken" links from Twitch's support center that didn't lead anywhere. A simple answer would suffice in the future.

Does anyone know how to go about submitting cover art for games on Twitch? A lot of the games that I follow just have that generic purple box with the Twitch logo in the middle and I'd like to give a few of them a proper cover.

American McGee's Alice single-segment speedrun or riot SwiftRage

I browsed the forums to see if this topic had previously been discussed and did not see any threads concerning the matter. My dilemma is that I run a game page for a leaderboard that is horribly incomplete. A lot of the runners have either moved on from speed running or at the very least, have moved on from running the game. However, I'm not sure if it would be considered unethical to add PBs for other players. I've seen this occur on countless leaderboards on this site and I don't really have any particular qualms with it (unless a specific runner has made it clear that they do not want their run on a leaderboard).

I'm not asking for the purpose of creating some philosophical debate. Instead, I simply would like to gain a better understanding on whether this is considered inappropriate to do. I want to see a complete and accurate leaderboard when possible but unfortunately, doing so would mean adding runs from players who may or may not want their runs posted.

What is the general consensus regarding posting other people's times?

So Becored found a way to kill the stingray roughly 2 seconds faster on the Japanese version, making it so that the Japanese version has an advantage of roughly 1:20. Since I don't have an exact time difference, I don't think I'm going to be able to consolidate the two versions into one category. Ideally, I would love to have just one table that lists someone's English time, the equivalent Japanese time (or the other way around, depending on which version your PB is on), and which version you used. I don't think I'm going to be able to do so now unless anyone has any ideas. Thoughts?

Here's a link to Becored's video:

Yeah I probably should have mentioned that. You're right, the pattern is different for the stingray. I kill the stingray in 4 cycles on both versions and while I've never actually taken the time to time my fight in both versions, I think it's virtually identical. I'm glad you mentioned that though. I'll try and time them both later on tonight.

Whether or not the slightly different stingray fight is worth making two separate categories I suppose is debatable. As I said before, I'm open to any and all suggestions concerning this game. Thank-you for your response. :)

EDIT: So I went ahead and timed my stingray fight from my English run and my Japanese run. I timed from when the room is first visible (since it fades black between the spike room and the stingray room) and stopped the timer at loss of character control. The first time I timed, I got 19.7 for BOTH versions. The second time I timed, I got 19.8 English and 19.4 Japanese.

I didn't bother to time it a third time because as you can see from the results, it's not particularly important. Granted, the way that you kill the stingray is going to have an effect on your time (obviously) but by killing him in 4 cycles in either version, the timing should be identical down to the second. Because I never advocate simply taking someone's word for it, I encourage others to test it for themselves and make sure that they get similar results. I hope this clarifies it.

Any suggestions on how to approach this leaderboard? The NES and Famicom versions are identical in gameplay but the Famicom version runs 1:18 faster due to less cutscene time. I'd hate to break the game into two separate categories since the gameplay is identical but I might have to do that. Listing the NES version as "time with loads" and the Famicom version as "time without loads" seems misleading and could be confusing to all but the few active runners of this game. I'm willing to listen to any and all suggestions before I edit the game on here.