This is a 2-player TAS of the snes game Super Bomberman 3,inspired by some youtube's tas which used one player and cheated(mine is faster hehe)
No passwords used, no death used, takes damage to save time, goes for pure speed
Movement and strategies are extremely optimized,but raw luck manipulation was used without any knowledge of the game's internal code(no scripts) so inputs were tested. This game's areas are beaten by destroying the orange lamps,so stage layout is manipulated to make those more acessible to a bomb.
The first world start slow because the items that can skip most of the fighting only appear later in the game and we're stuck with blue "yoshi" to kick bombs all over the place so i can destroy the lamps fast. Many times,the soft block layout used makes one character useless but i made sure this was actually the fastest way to beat the area. Also,items are fixed per stage but their position on it is not.
Interesting stages to comment:
1-3 is a horrible stage,i couldn't get anything better for the layout and blue yoshi's kick is revealed to actually restart the bomb's timer here!
1-4 is mostly the same thing happening :( In its second area,i discovered that i cannot put the bomb in the first frame with the second character for some weird reason,hence why i end up putting the bomb in the next block.No time seems to be lost cause the enemy would have blocked the bombkick if it had gotten there sooner.
1st boss battle is perfect.I had found a glitch with becoming a bunny in the end of the fight mixing the color pallette,but it costed time so it was not used.
2-3 had good stage layouts but the final area also had the remote bomb in a bad position(all possible layouts tested in the first 7 frames had it so i kept this one for good lamp position)
As soon as i get the remote bomb,i start abusing invincinbility to detonate bombs inside other bomb's explosions.
2nd boss was good but i feel it can be improved by getting closer to the starting position in the end of the fight.Some bosses put you automatically going there before completing the death cutscene,which is why i always go back there.
3-2 i get the glove and use the blue "yoshi" as shield so i can detonate the bomb sooner,because the glove is a faster method fo throwing bombs around the screen(yoshi needs to take a few steps aways from the bomb before kickin it).
3-3 more remote bombs and the blockpass will cut this game's time i'd say in half.
3rd boss destroys me,i haven't found a way to repeat the attack which permits hitting twice many times in a row,so this guy actually gets vulnerable before he can be hit.Only the head and the neck are "hittable".
From here on it's a smooth ride,even though there is not much space to play around cause my inputs advance RNG for the next area's layout. 4th boss goes perfect,5th world and bosses too,6th world is best cause i spend no frame manipulating most areas.
final boss is ok in the first part,could be better if he could be manipulated to stay in the center(transformation for the second part begins when bomberman and bagular are positioned in their starting locations)
lol the run is not deathless,i wanted to show that you can stand between the robot's legs but it wasn't faster at any moment to put the bomb there cause it triggered attacks that would be very compromising in terms of time to dodge.
hope you enjoyed,i will answer more specific question in thread if necessary,and i believe all future improvements will come from better RNG understanding for stage layouts and the 2nd,3rd and final boss behaviors.Movement and strats was very tightly optimized so i'm pretty satisfied with this.The movie was tiring to do because the emulator is still very young(i haven't found a input editor version of it) but it has my full support cause it's very good.
 signin off and thankin you for reading all this,my fellow players for being patient and helpful,see you all in the forums.

Nach: Judging.
Ilari: Replaced savestate with movie (no input changes).

Nach: Reviewing the run, for the most part, it looks good. A few segments are questionable but aren't too bad. Regarding some levels being slower than "random encodes", it is unjust to pit runs which use passwords, inferior emulators, and are simply video which cannot be tested, against a run made with a proper emulator with a proper starting point and running process which is known to be clean and fair. True, there is what to improve, but for a first run for this game, I find this acceptable.
Ilari: Processing...

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14852
Location: 127.0.0.1
This topic is for the purpose of discussing #3910: KusogeMan's SNES Super Bomberman 3 in 25:45.97
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
http://www.youtube.com/watch?v=fXO6oEMl0RY&feature=youtu.be can i vote yes on my own TAS?lol the link above is a low quality encode.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 5/12/2009
Posts: 748
Location: Brazil
Did you find it entertaining? :)
Joined: 3/9/2009
Posts: 530
Are you sure this is optimized? Example: 32 seconds in, P1 both stops moving for a moment, then places a bomb at a crossroad instead of in the alcove and then has to wait 2 extra spaces away for the explosion to finish. Both of those look really blatantly unoptimal and there are more small things as it goes on. Edit: Single player is only marginally slower than you on that stage too, and about 7 seconds faster than you on stage 3. http://www.youtube.com/watch?v=suEeI5vJd0U
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
Tangent wrote:
Are you sure this is optimized? Example: 32 seconds in, P1 both stops moving for a moment, then places a bomb at a crossroad instead of in the alcove and then has to wait further 2 extra spaces away for the explosion to finish. Both of those look really blatantly unoptimal and there are more small things as it goes on. Edit: Single player is only marginally slower than you on that stage too, and over 5 seconds faster than you on stage 3. http://www.youtube.com/watch?v=suEeI5vJd0U
i thought this was the cheated run but it's not...stage 3 really went bad. i can't really figure out what part exactly what you're talkin about but at 32 seconds p1 needs to explode the orange lamp as soon as possible,hence why it puts on the crossroad,regardless of distance he is from it.p2 waits to kick the bomb while the fire is active,too soon it bounces,too late and i lose frames waiting for it explode.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 3/9/2009
Posts: 530
grassini wrote:
Tangent wrote:
Are you sure this is optimized? Example: 32 seconds in, P1 both stops moving for a moment, then places a bomb at a crossroad instead of in the alcove and then has to wait further 2 extra spaces away for the explosion to finish. Both of those look really blatantly unoptimal and there are more small things as it goes on. Edit: Single player is only marginally slower than you on that stage too, and over 5 seconds faster than you on stage 3. http://www.youtube.com/watch?v=suEeI5vJd0U
i thought this was the cheated run but it's not...stage 3 really went bad. i can't really figure out what part exactly what you're talkin about but at 32 seconds p1 needs to explode the orange lamp as soon as possible,hence why it puts on the crossroad,regardless of distance he is from it.
You stop moving for a moment when you place THAT bomb, and the one after that should be not at the crossroads because it forces you back extra spaces. Jumped to 10 minutes in and the remote clearly wasn't detonated as quickly as it could have been. Spot checked another comparison. http://www.youtube.com/watch?v=QM1Ki7WcbBQ&#t=6m28s http://www.youtube.com/watch?v=fXO6oEMl0RY&#t=15m28s Single player is faster than yours again. You should look over his run and make sure that you're beating or at the very absolute least matching those times.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
Tangent wrote:
grassini wrote:
Tangent wrote:
Are you sure this is optimized? Example: 32 seconds in, P1 both stops moving for a moment, then places a bomb at a crossroad instead of in the alcove and then has to wait further 2 extra spaces away for the explosion to finish. Both of those look really blatantly unoptimal and there are more small things as it goes on. Edit: Single player is only marginally slower than you on that stage too, and over 5 seconds faster than you on stage 3. http://www.youtube.com/watch?v=suEeI5vJd0U
i thought this was the cheated run but it's not...stage 3 really went bad. i can't really figure out what part exactly what you're talkin about but at 32 seconds p1 needs to explode the orange lamp as soon as possible,hence why it puts on the crossroad,regardless of distance he is from it.
You stop moving for a moment when you place THAT bomb, and the one after that should be not at the crossroads because it forces you back extra spaces. Jumped to 10 minutes in and the remote clearly wasn't detonated as quickly as it could have been. Spot checked another comparison. http://www.youtube.com/watch?v=QM1Ki7WcbBQ&#t=6m28s http://www.youtube.com/watch?v=fXO6oEMl0RY&#t=15m28s Single player is faster than yours again. You should look over his run and make sure that you're beating or at the very absolute least matching those times.
1my overall time is faster,isn't it? i'm not saying mine is always faster at all stages... 2the remote bombs don't need to be detonated as fast as possible if you're not blocked by a soft block,it's the hit on the last orange lamps that have to be optimized.They might LOOK LIKE they are detonated earlier because his character moves faster,but i can assure i use the inputs A,skipframe,B to detonate my bombs all the time.This is something you can see in the movie file. 3the time that the player gains might be the snowball effect of bomberman games where one character doesn't share his items and is,therefore,stronger and faster than the ones that split items but can't cover all the places at once as a team would do.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Skilled player (1705)
Joined: 9/17/2009
Posts: 4952
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
grassini wrote:
1my overall time is faster,isn't it? i'm not saying mine is always faster at all stages...
But it has to beat all existing records, not just overall.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
but my TAS is faster than the one presented,some of the stages are slower,but not all. Is there another TAS which i'm not aware of?
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Skilled player (1705)
Joined: 9/17/2009
Posts: 4952
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
grassini wrote:
but my TAS is faster than the one presented,some of the stages are slower,but not all. Is there another TAS which i'm not aware of?
No, but doesn't existing records imply individual stages in an existing run? Unless you have a reason like "RNG, Frame Rule, Speed/Entertainment tradeoff, etc.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
jlun2 wrote:
grassini wrote:
but my TAS is faster than the one presented,some of the stages are slower,but not all. Is there another TAS which i'm not aware of?
No, but doesn't existing records imply individual stages in an existing run? Unless you have a reason like "RNG, Frame Rule, Speed/Entertainment tradeoff, etc.
The layout is random!I'm not sure how many frames i'll have to wait to get the exact layout that would give me the same strategy to "tie" these individual stages.This is why i listed the improvements as better luck manipulation.Also,i wouldn't even be able to use the same strategies as the item distribution for 2 players would be totally different.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 3/9/2009
Posts: 530
grassini wrote:
jlun2 wrote:
grassini wrote:
but my TAS is faster than the one presented,some of the stages are slower,but not all. Is there another TAS which i'm not aware of?
No, but doesn't existing records imply individual stages in an existing run? Unless you have a reason like "RNG, Frame Rule, Speed/Entertainment tradeoff, etc.
The layout is random!I'm not sure how many frames i'll have to wait to get the exact layout that would give me the same strategy to "tie" these individual stages.This is why i listed the improvements as better luck manipulation.Also,i wouldn't even be able to use the same strategies as the item distribution for 2 players would be totally different.
You also have twice as many characters doing the same amount of work. You should be faster, and easily observably faster much of the time. Certainly not 7 seconds slower than a single player in stages, especially that late in the game when you've had plenty of time to get your items set up.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14852
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [2355] SNES Super Bomberman 3 "2 players" by KusogeMan in 25:45.97
Player (98)
Joined: 3/20/2008
Posts: 466
Location: Montreal, Quebec, Canada
Why would you publish a run with 2 yes votes and 2 no votes? In addition, the only feedback for the run are highly critical posts by Tangent. So not only is there clearly insufficient feedback on which to base a decision, the existing feedback is negative enough to seriously doubt whether this run is publication worthy.
Joined: 3/9/2009
Posts: 530
Vykan12 wrote:
Why would you publish a run with 2 yes votes and 2 no votes? In addition, the only feedback for the run are highly critical posts by Tangent. So not only is there clearly insufficient feedback on which to base a decision, the existing feedback is negative enough to seriously doubt whether this run is publication worthy.
I seriously wasn't expecting this to go through at all. If you need some patently obvious evidence that it's extremely suboptimal and didn't have any planning or research into it... http://www.gamefaqs.com/snes/564897-super-bomberman-3/faqs/32363 Items are hard set by level, location is random. Checked it for 2 player too. Distribution is identical. Which means an excrutiatingly obvious time saver is picking up the skates (increased walking speed, up to 9 total) in stages 1-1 and stages 1-3, which this run doesn't do. He also misses the skates in 3-3 (was checking to make sure he got block pass). Missing/skipping speed up items that early in the game is... Edit: For giggles, I also checked 2-2, 2-4, and 4-1. Doesn't pick up the skates in any of them.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
Tangent wrote:
Vykan12 wrote:
Why would you publish a run with 2 yes votes and 2 no votes? In addition, the only feedback for the run are highly critical posts by Tangent. So not only is there clearly insufficient feedback on which to base a decision, the existing feedback is negative enough to seriously doubt whether this run is publication worthy.
I seriously wasn't expecting this to go through at all. If you need more clear evidence that it's extremely suboptimal without understanding the game's mechanics... http://www.gamefaqs.com/snes/564897-super-bomberman-3/faqs/32363 Items are hard set by level, location is random. Checked it for 2 player too. Distribution is identical. Which means an excrutiatingly obvious time saver is picking up the skates (increased walking speed, up to 9 total) in stages 1-1 and stages 1-3, which this run doesn't do. He also misses the skates in 3-3 (was checking to make sure he got block pass). Missing/skipping speed up items that early in the game is...
We've discussed this already.To be honest,i'm also surprised to see it published so fast given the fact that i'm not sure whether the items can be found amidst the fastest layout,but then again,nor is Tangent.I know the items are hardset but i either manipulate for items or for good stage layout. The real disagreement here lies in the fact that while i made my TAS with the idea that the items i would get would have to be on the way of the less destructive path,Tangent believes i have to get out of my way or somehow manipulate the items to be in different places with these same or similar best layouts,and that the time lost for getting an item would obviously be recovered later with the frames gained with sheer walk speed.(If items had been in the way of these layouts or similar ones,the TAS would surely be faster but can it be done?I had to explode each stage's many layouts completely to take notes of item locations) None of us can prove that it's completely impossible or possible to do that. What I do know is that i couldn't manage to switch the item positions without changing the layout,no button combination or simply waiting gave me a result with different item position(and i managed to cycle through possibilities a few times in a few stages).Someone who is good with memory adresses and scripting should be able to prove me right or wrong,because no judgement without it is better than anybody's else judgement.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 3/9/2009
Posts: 530
grassini wrote:
None of us can prove that it's completely impossible or possible to do that. What I do know is that i couldn't manage to switch the item positions without changing the layout,no button combination or simply waiting gave me a result with different item position(and i managed to cycle through possibilities a few times in a few stages).Someone who is good with memory adresses and scripting should be able to prove me right or wrong,because no judgement without it is better than anybody's else judgement.
http://www.youtube.com/watch?v=suEeI5vJd0U Here you go. Here's five year old proof that it can be done with half as many characters. Note that at the end of the first world, you're BEHIND the single player by about 3 seconds AND your characters don't have as many power ups as his, despite 1p only being able to cover about half as much ground more or less AND him taking extra time to set it to HARD (not to mention in BEING on Hard). Edit: I'm also a little curious as to why you accused that run of cheating.
Joined: 3/9/2009
Posts: 530
I hate to bump this, but this is more directed to Nach/the judging decision and comments
it is unjust to pit runs which use passwords, inferior emulators, and are simply video which cannot be tested, against a run made with a proper emulator with a proper starting point and running process which is known to be clean and fair
That run is five years old so bagging on it for emulator choice seems really weird, the password used is to set it to Hard mode (enemies have more health), and 4432 is an old Japanese TASer mainly known for his Kirby stuff, so insinuating that he cheated it somehow is pretty iffy. Rejecting it as a valid comparison for those reasons is pretty faulty or outright wrong in the case of the password.
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
I know it's not cheated,i confused it with another TAS i had seen of bomberman 3.I have said this in my comments up there. Still,I think this run is publishable on the basis that it is the best offered on the site's standards even though i would love to see it obsoleted by a run with a thorough understanding of the game's RNG. Tangent,don't be shy to prove me wrong by TASing some bomberman 3 yourself.It would make me happy to know this disagreement can foster competition,improvements and maybe cooperation.
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general
Joined: 3/9/2009
Posts: 530
grassini wrote:
Still,I think this run is publishable on the basis that it is the best offered on the site's standards
Actually, I've shown that it badly fails to meet technical muster to which you have given no answer except declaring RNG manipulation, a normally integral part of TASing, to be beyond you, and once we take your vote out, it has only a 25% approval rating, so... Which standards does it exemplify again?
SDR
Skilled player (1360)
Joined: 4/1/2009
Posts: 245
Location: Countryside of anywhere
wow there's a big discuss here about this run but I'm not going to start a fight, instead of that, I'd like congratulate you by your first run accepted here. Congratulation
whatever worth doing, is worth doing well
Active player (434)
Joined: 2/5/2012
Posts: 1687
Location: Brasil
thanks,i appreciate a lot!
TAS i'm interested: megaman series: mmbn1 all chips, mmx3 any% psx glitched fighting games with speed goals in general