The TAS doesn't actually throw the last mecha-koopa. They end timing before the last mecha-koopa is even thrown by Bowser. Technically this strat loses real time compared to RTA because Mario takes damage and the damage animation costs a bunch of frames, but for TAS timing it doesn't matter because the movie has already ended by this point.
This alone should be reason to not compare TAS to RTA. idk why people care whether RTA wins or not. RTAs are a skill showcase, TASes are for entertaining an audience. They have different rules, different goals, and even different communities.
Because, since the RTA being played by imperfect humans, it should not be faster than the supposed "perfect play" of a TAS. And it is trivial to compare different timings: the RTA community does it all the time.
One of the rules on this site is that a TAS must beat ALL RECORDS, including any RTA records. If it isn't doing so, that is a problem.
Comparing the two is required by the site rules.
[
One of the rules on this site is that a TAS must beat ALL RECORDS, including any RTA records. If it isn't doing so, that is a problem.
The SMW TAS community consider the ACE TAS as a TAS of the RTA any%‚ as they both use the cloud. Some people don't agree‚ but unless new people start TASing the game, you won't get an optimized "warps" TAS using the cloud.
I problably made mistakes, sorry for my bad English, I'm French :v
[
One of the rules on this site is that a TAS must beat ALL RECORDS, including any RTA records. If it isn't doing so, that is a problem.
The SMW TAS community consider the ACE TAS as a TAS of the RTA any%‚ as they both use the cloud. Some people don't agree‚ but unless new people start TASing the game, you won't get an optimized "warps" TAS using the cloud.
And if someone makes one, the argument over whether it should be accepted will last 10 pages :)
Joined: 11/13/2006
Posts: 2823
Location: Northern California
trlky wrote:
Because, since the RTA being played by imperfect humans, it should not be faster than the supposed "perfect play" of a TAS. And it is trivial to compare different timings: the RTA community does it all the time.
One of the rules on this site is that a TAS must beat ALL RECORDS, including any RTA records. If it isn't doing so, that is a problem.
Comparing the two is required by the site rules.
Are you Dyshonest's alternate account or do you just have similar troubles grasping extremely simple concepts?
If you actually watch any recent RTA run and the ACE run, you'll see that the beginning of both runs is exactly the same. The same glitch (chuck-eat glitch) is used with the same setup (though the TAS does it much flashier, obviously). Even without any further explanation it should be obvious to anyone that the TAS uses the same glitch to a much, much better effect than the RTA is able to.
This particular application of the chuck-eat glitch crashes the game, giving a small window of opportunity to send input to open bus. Without the right set of input, the game stays crashed and the run is lost. Sending input to open bus means that the game's code is directly edited: Open bus is not meant to be directly accessible via player input, so anything that gets sent to it is executed immediately as it's not expecting anything except what it was originally programmed for. Therefore: Both runs use ACE. The RTA run scrolls the screen at that moment, as it's the most consistent way to prevent the game from crashing completely (though it still does crash completely fairly often). This gives the impression that it's not a major game-breaking glitch in any way, especially since chuck-eat has been used in other SMW runs without any extra effects (namely, getting the goal orb to skip Iggy's Castle, which doesn't crash the game to open bus at all).
It's not complicated. You don't need to have any coding knowledge or SMW TASing knowledge to understand it. All my information comes from the 20 times this has been discussed before, that you should have read before presenting your uneducated opinion.
If you really want to see an "update" to the warps TAS that uses the cloud, go ahead and make it yourself. Don't worry about optimizing it for submission to the site: It won't be accepted no matter how optimized it is.
TASvideos Admin and acting Senior Judge 💙 Currently unable to dedicate a lot of time to the site, taking care of family.
Now infrequently posting on BlueskywarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
I wrote a lua script of Super Mario World for lsnes. The lastest (stable) version can always be found here:
https://github.com/rodamaral/smw-tas/blob/master/smw-lsnes.lua(last edited 2014/09/08)
This script works for lsnes stable rr1 and lsnes beta rr2. Let me know if it has bugs and don't be shy to suggest improvements. Some observations:
1) The script will try to detect the version of Lsnes that you're using, but you can change that in the options. The beta version has more features that doesn't work in the stable version. For instance, there's no lag count or lag flag in rr1 and you have to use the lateral status panel to see the lag.
2) Mouse right click: that makes the hitbox of a sprite appear (or disappear). The mouse must be near the sprite. Also works with Mario.
3) Mouse left click: that makes the hitbox of a block appear (or disappear).
4) Button "-" (minus) diminishes the transparency of the script. Button "=" (equal) increases it.
5) The hitbox of the enemies are not accurate yet. Only some of them are and I will improve it soon.
6) Mario's hitbox includes 2 kinds: interaction with blocks (white points) and interaction with sprites (red lines). I changed that approach very recently, so tell me if that in unintuitive.
7) There's a cheat, enabled by default, to beat the current level.
Start + Select + X: activates the normal exit
Start + Select + A: activates the secret exit (don't use it in levels without it)
Start + Select + B: simply exits the level
Is there a table which shows what powerup you get at each PI number?
I didn't understand your question, can you elaborate more?
You can use the PI glitch to get 256 different values, right?
Has everyone tested each of these values, documented the effects and made a list of it all?
I went through all 255 items you can get in your inventory box and found some rather interesting ones that affect bowser, besides cloud:
Link to videoLink to video
It's worth noting that both the Big Boo boss(as found by Volpey) and Lemmy/Roy will trigger credits when killed in any room of the castle.
For the record the discovery above got put on reddit.
http://www.reddit.com/r/speedrun/comments/2qg213/super_mario_world_new_bowser_glitches/
My 2 cents: I'm sure if you guys went through and tried eating all kinds of chucks, but I think the idea behind the glitch should be looked at. Perhaps eating another enemy or object such as a blargg, boo, block would give a greater item id pool.
There has already been a Wrong warp to credits done by JeffW in 4:59
I don't see how eating any other chuck/object would save any time in a any% run unless you can get it in DP1, in a 96 run any of these glitches would be banned, and as for TASing goes, they would just get rejected as they can already use ACE to warp to the credits in around 40 seconds.
Can always use the PI glitch to get anything you want in the item in Reserve in YI3.
ah so the yellow switch inside the palace is not a sprite but an object?
In fact, you can't eat the switch, but you could (theoretically, via item swap, but seems impossible) eat the invisible block that is present in the room. Sprite 0x6d is inedible but it curiously has the "give powerup when eaten by Yoshi" flag, just like the Chucks.
I have tested a little bit now, but only crashed the game.
EDIT: something good can come out of this. Eating it small or with the first PI after the flower gives a feather in the item box.
ah so the yellow switch inside the palace is not a sprite but an object?
In fact, you can't eat the switch, but you could (theoretically, via item swap, but seems impossible) eat the invisible block that is present in the room. Sprite 0x6d is inedible but it curiously has the "give powerup when eaten by Yoshi" flag, just like the Chucks.
I have tested a little bit now, but only crashed the game.
EDIT: something good can come out of this. Eating it small or with the first PI after the flower gives a feather in the item box.
yeah this was what i was looking for i mean you can't jump on something without triggering something but i guess its a ded end
For the record the discovery above got put on reddit.
http://www.reddit.com/r/speedrun/comments/2qg213/super_mario_world_new_bowser_glitches/
My 2 cents: I'm sure if you guys went through and tried eating all kinds of chucks, but I think the idea behind the glitch should be looked at. Perhaps eating another enemy or object such as a blargg, boo, block would give a greater item id pool.
The whole reason the chucks work is because they are accidentally coded to give Mario a powerup when Yoshi eats them. If the latter three aren't coded this way, then sprite-swapping them probably wouldn't have any effect.
Hello! I am DrTAS! I have a TAS idea in which I have already started on and would like opinions on it before I do more. The idea is (I know this might sound crazy) that I will focus on doing the powerup increment glitch and after I do the glitch twice I go to the next level in the game. I will keep on doing this until powerup 106. Hence the name "PI 106." So you probably know what PI 106 is so I don't have to explain but if you don't know then look it up on google. This TAS will be long and boring but I hope the ending is cool. I would like as much feedback as possible. Thanks!
I happen to still have two .smv files from when we were still on the project more than 4 years ago. This was all my input I believe.
http://dehacked.2y.net/microstorage.php/info/1498898640/YW%20Redo.smv - From power on, I did the PI up until halfway through the turn block duplications of YI3.
http://dehacked.2y.net/microstorage.php/info/1284079316/yi3-afteryw.smv - I did PI 3 times inside YI3 after completing it once. The goal was this was being able to copy-paste the input the amount of times needed to reach PI 106.
Honestly, if enough attention is gathered, perhaps this project could be revived, although obvious objections would be the entertainment value from PI levels 5-106, as well as skipping almost all levels. And I don't think SMW needs yet another branch category (glitched 100%). Perhaps could be a vault TAS.
Vault wouldn't accept this kind of TAS iirc; it only allows pure any% and pure 100%. "Special" goals have to be entertaining enough for Moons or else rejected.On second thought, this is technically the fastest possible way to get 100%, although more boring than the non-glitched way. So this might just be exactly the kind of thing Vault is for.
On third thought, arbitrary code execution is unarguably the fastest way to do everything, so a PI glitched 100% run would be obsoleted by that. But there have also been special rules in the past that automatically ban ACE for non-any% runs by default.
...Any help from someone more knowledgeable about the rules?
A question here i know that PI106 gives you the ability to finnish a level with regular goal and "secret" goal and i know you can spawn a key/keyhole on Funky to get an additional exit in your title screen on J version. how would it affect to use the PI106 secret exit on every level that doesn't normally have a secret exit?
Hello guys! It is me, puppy610 and I would like to ask a few questions about ACE for Super Mario World.
1) Can someone give me like a list of what controller inputs do what commands?
2) Is it possible to use ACE in SMW and edit the value that shows how many exits you have in the title screen and if so what are the ACE program inputs?
3) Can I make ACE execute the end of a level instantly for any level in SMW? If so please give inputs.
Thank you so much! Have a nice day!
