This run improves the current published one by 645 frames (~10.7 seconds)

Aglar's Comments

Finally we have produced something! We started this in late 2009 and thought it'd be a rather small project, but that turned out to be not exactly true and it has now basically taken much longer than both of us or anyone in the community who has sort of followed this could've imagined - andit's not over yet. After some time we decided to do both this run and the warpless one, with the intension of working on them simultaneously and submitting them at the same time when done. This was something that we had to give up in the end though. The warpless run gave us so much trouble that we decided to release the warped run first as opposed to letting all you wait an unknown amount of time for both runs.
I think this started to become a huge project when I found that defeating Birdo with a shell saved time in level 1-1, something that further evolved into even better strategies in that level and also opened our eyes for the possibility that this game might have quite some unexplored secrets. Since then we've found numerus of new tricks and strategies, mostly for the warpless run though so you won't see them here:).
I was mostly the one who took care of the optimization part of the run as I quickly got a good hang of how the subpixels work and in the end I think all the input in this run are from me. However, without Andrewg the run wouldn't have been close to as fast as it ended up being (the final strategy used in level 1-1 was his idea which i'd never thought about) and his discoveries play an even bigger role in the warpless run. I think we complemented each other really well!
I'd like to thank all the previous autors of this game who've pushed it to such a great standard that the only way of keep up with it is learn very well how the game works, just being lucky won't work. I also thanks Glitchman for finding that you can fall through the floor using a vegetable. I also thank the ones the Andrewg thanks (He wrote his comments before me).

Andrewg's Comments

I had a really fun time working on the Super Mario Bros. 2 TAS. As I currently hold the record console time for the game, it was interesting to discover new tricks and glitches in the game. Aglar is a great guy to TAS with, his optimization skills are far beyond mine and he's a really creative guy.
Together, we managed what I'd say is far beyond either of our initial expectations for a TAS of this game. I thought maybe we would save a just time in level 1-2 with the discovery of a new glitch - the "carpet glitch", which was found to save time, discovered by CtrlAltDestroy. He and Guanobowl actually mentioned several things that we were able to duplicate and or just find with complete luck. The fact that we saved time in nearly every level shows the amount of time and effort we spent with this. The real rerecord count for this TAS is probably close to 200k.
I was also really curious of all the glitches mentioned in the Mario 2 thread when I first started this TAS, so I set out to see if I could "solve" each of them. I think I figured out all that were mentioned now, or at least how to duplicate them or use them to our advantage.
Anyway, I had a lot of fun with this. 1-1 especially was a level where I enjoyed experimenting and testing my crazy ideas. :) Somehow, my ideas actually worked!

People I'd like to thank:

- Adelikat for initially helping me with understand how this game worked a bit better and sending me more info on the game.
- CtrlAltDestroy & Guanobowl - as I actually ended up doing more experiments because of all the tricks and glitches you guys mentioned in the Mario 2 thread.
- Previous runners - for various info I gained reading through submissions and watching older TASes, especially "Sleepz" whom seemed to find quite a few improvements in 2004 or so.
- Aglar for putting up with my procrastination on both the warp and warpless runs. I got very frustrated at points because optimizing in this game is extremely difficult.

Frames saved level by level

LevelFrames savedTotal frames saved
1-1461461
1-2129590
4-22592
6-14596
6-342638
7-11639
7-26645

Level-by-level comments

1-1: Huge time saver, and if you checked the fm2 that was posted in the game thread last year you've already seen this - or have you?
1-2: Used the carpet glitch that most people know about by now, it ended up saving 119 frames. The Birdo fight was done 10 frames faster.
4-2: The vine glitch resulted in this small improvement.
6-1: Tight subpixel management saved 1 framed going down the platforms before the sand digging area, and 3 frames were saved from jumping up the sand area better.
6-3: New strategy for Triclyde saved a bunch of frames.
7-1: Lag reduction.
7-2: Lag reduction and throwing the last vegetable at Wart better.

New glitches used in the run (spoiler alert)

Carpet glitch in 1-2:

If there are two carpets on screen at once, it is possible to make them move at extremely fast speeds.

Carry an enemy to the next room:

Drop the POW-block and pick up an enemy just as the block hits the ground, you know it works if you get the "pick up" animation without picking up the enemy, then enter a new room and you'll be holding it.

Getting pushed down while on a vine/ladder (in this run, ladder):

Normally if you're on a ladder and a bomb falls down on you, you'll catch it while still climbing, however if you're already holding a bomb as in this run and another one falls at you it'll push you down the ladder.

Vine glitch in 4-2:

If you're on the very left side of the vine when entering a new screen you'll start a little more to the right in this screen.

DarkKobold Claiming for judging
Flygon: Corrected YouTube module size to the agreed upon standard for NES games.
GabCM: Added comparison link
DarkKobold: This is a no-brainer. New glitches, slight optimizations elsewhere... Obvious accept.

BigBoct
He/Him
Editor, Former player
Joined: 8/9/2007
Posts: 1692
Location: Tiffin/Republic, OH
Bisqwit wrote:
Co^gratulawionq$oѮ!anotmeɢࠠgɡme䀠fcnhshbd sušCjgr`}ian ѡŮyo~a aowll `ave gtɵၷsed! 
What the hell?
Previous Name: boct1584
Joined: 5/15/2006
Posts: 102
Holy motherfuxxing trees with hot dicks from slimy hell!!! This is one of the greatest improvements ever. Some people even went crazy after watching it, like Bisqwit. :)
Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
1ntru wrote:
Holy motherfuxxing trees with hot dicks from slimy hell!!!
I'm generally against censorship, but let's keep it more tasteful than that.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Joined: 3/22/2010
Posts: 6
Location: London
Fantastic TAS. Quick question: has it happened before that the speedrunning record and the TAS record were held by the same person? (I get the vibe Adelikat is one of those people.)
SITS
Senior Moderator
Joined: 8/4/2005
Posts: 5770
Location: Away
It has, many times. Mike McKenzie (mike89) has had a record in Vectorman 2 any%, Kirk LaBuda (BioSpark) has had records on Metroid Fusion, Robin Reigstad (Deign) has had records on Final Fantasy IV, inichi has had a record on Chrono Trigger any%. AndrewG, Seth Glass, and stanski have had unassisted and TAS records for the same game held simultaneously as well, although the TAS records were shared between multiple players. There are definitely more examples. Many of the early contributors were unassisted speedrunners (Josh the FunkDOC, Sleepz, stx-Vile, LLCoolDave, SprintGod, Gigafrost to name a few), so it wouldn't suprise me if the absolute count was even higher in the beginning.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Joined: 5/15/2006
Posts: 102
moozooh wrote:
1ntru wrote:
Holy motherfuxxing trees with hot dicks from slimy hell!!!
I'm generally against censorship, but let's keep it more tasteful than that.
прости меня. I just thought that this would be some sort of appropriate way to put this TAS into words. Epic or amazing won't do.
Former player
Joined: 7/4/2005
Posts: 714
Location: Albuqueruqe, New Mexico
Freakin' nice boss. Thumbs up from me. Also, thanks for the mention in the description. Yes vote
Change my sig. again, and I will murder your pet fish.
Experienced player (699)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
You will notice that with almost every game I decide to TAS, I also have the unassisted speedrun record. I generally TAS the game I'm currently doing a console speedrun of ;) hint hint. It's just the way it works out. If I find a cool new game, I like to TAS it and speedrun it together. (Ghoul School and Werewolf) Also, thanks for all the complements. This TAS took quite a bit of work. I did a ton of testing and Aglar's optimizations are insane.
Super Mario Bros. console speedrunner - Andrew Gardikis
Banned User
Joined: 6/18/2010
Posts: 183
moozooh wrote:
1ntru wrote:
Holy motherfuxxing trees with hot dicks from slimy hell!!!
I'm generally against censorship, but let's keep it more tasteful than that.
Doesn't that really depend on the tree?
Banned User, Player (142)
Joined: 8/30/2010
Posts: 500
Location: Argentina Bs. As.
[18:51] <scrimpy> Oh, nothing [18:51] <mmarks> oh [18:51] <Nach> I think scrimpy is just jealous of you mmarks
Noxxa
They/Them
Moderator, Expert player (4138)
Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
mmarks wrote:
Mister Epic wrote:
Click here to download a lossless, SD and full frame rate comparison video of this and the movie to be obsoleted.
Not Found?
The link is broken due to quotation marks in the [url] tags. Fixed link: http://www.mediafire.com/?z1yambw1as2coiy
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
GabCM
He/Him
Joined: 5/5/2009
Posts: 901
Location: QC, Canada
Sorry for that mistake. Go to the link in Moth's post!
Player (79)
Joined: 2/8/2005
Posts: 130
My favorite of the Mario runs
Sir_VG
He/Him
Player (39)
Joined: 10/9/2004
Posts: 1911
Location: Floating Tower
Interesting how the new run was behind a smidge during 8-2 and still wound up 6 frames ahead in the end.
Taking over the world, one game at a time. Currently TASing: Nothing
Editor, Experienced player (608)
Joined: 11/8/2010
Posts: 4012
Nice side-by-side encode, Mister Epic! It was cool to compare the frames as the movies were both playing and it was interesting to see the size of the improvements. As others have said, I also look forward to the improved warpless run.
Mr. Kelly R. Flewin wrote:
To no surprise, someone voted no... definitely curious as to why as this run had EVERYTHING!
I'm curious too. But if this submission gets 200 Yes votes, the percentage of No votes should round down to 0%. (I don't think there will be that many votes, but it would be nice to see.)
Active player (458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
This run was very entertaining and I hate SMB2 with a passion :O. Almost wanted to hook-up the NES and play it. Almost. Anyways, thanks for making this run :D. Strong yes vote.
Noxxa
They/Them
Moderator, Expert player (4138)
Joined: 8/14/2009
Posts: 4083
Location: The Netherlands
CoolKirby wrote:
Mr. Kelly R. Flewin wrote:
To no surprise, someone voted no... definitely curious as to why as this run had EVERYTHING!
I'm curious too. But if this submission gets 200 Yes votes, the percentage of No votes should round down to 0%. (I don't think there will be that many votes, but it would be nice to see.)
If I recall correctly, submission vote percentages always round down, so it would be 0% at 100 yes votes (101 votes total). That's a lot closer than 200.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
GabCM
He/Him
Joined: 5/5/2009
Posts: 901
Location: QC, Canada
Wow! More than 50 downloads for my comparison already! And good feedback as well! Thank you! I think I'll make more of them, and maybe an HD encode for this one, for my YouTube account. And also, a Yes vote from me. Nice improvement!
Joined: 7/2/2007
Posts: 3960
Comparison videos get painful to listen to if the sound gets badly out of sync. That only happened once in this run, fortunately. If you make a comparison video for a run where the music doesn't reset as often or where the improvements are more frequent and/or drastic, you might want to consider only pulling sound from one of the movies. It'd also be neat to show the per-segment improvement and the total improvement so far. That said, thanks for making the comparison vid! It was fun to watch.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
GabCM
He/Him
Joined: 5/5/2009
Posts: 901
Location: QC, Canada
Derakon wrote:
Comparison videos get painful to listen to if the sound gets badly out of sync. That only happened once in this run, fortunately. If you make a comparison video for a run where the music doesn't reset as often or where the improvements are more frequent and/or drastic, you might want to consider only pulling sound from one of the movies.
I'm thinking of pulling the sound of each movie on each stereo channel next time.
Derakon wrote:
It'd also be neat to show the per-segment improvement and the total improvement so far.
Yeah, I'll try that next time, but there is no guarantee I can make such a thing.
Derakon wrote:
That said, thanks for making the comparison vid! It was fun to watch.
No problem!
Banned User, Former player
Joined: 3/10/2004
Posts: 7698
Location: Finland
The comparison video is great, but it's annoyingly unseekable (or at least my version of mplayer cannot seek it). Probably lacks keyframes or something.
Joined: 7/2/2007
Posts: 3960
Mister Epic wrote:
Derakon wrote:
Comparison videos get painful to listen to if the sound gets badly out of sync. That only happened once in this run, fortunately. If you make a comparison video for a run where the music doesn't reset as often or where the improvements are more frequent and/or drastic, you might want to consider only pulling sound from one of the movies.
I'm thinking of pulling the sound of each movie on each stereo channel next time.
Ehh, while that sounds neat in practice it doesn't make much difference, at least for me. I'm still listening to two separate audio tracks overlaid on each other. I suppose it'd be possible to suppress playback of one of the channels, which you can't when they're not separated, but do media players actually support that option?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Player (142)
Joined: 7/16/2009
Posts: 686
It's possible to include multiple audio tracks to a video, so you could include three: One for both movies, one for the old movie and one for the new movie. Although I think it's also possible to give both movies one of the stereo channels, and then mute either of them.