This is my first post and this is my third run of quad-run.I refered to many movie files.So I need to say "Thanks" to authors of many movies.
Game Objectives
FCEUX 2.1.3-interim svn1599
Uses same roms as current fastest movie used
Aims for fastest time
Four games in one movie
Manipulates luck
Take damage to save time
No death
Abuses programming errors
Useful Information for the run
If you push left and right on the same time, SMB1 and SMB2J will walk toward right, SMB2 will run toward right and SMB3 will stop accelerating.Even if SMB3 jump, SMB3 won't accelerate.
If you push left or right and bottom, SMB1,SMB2J and SMB2 will stop accelerating unless SMB1,2J and 2 are jumping and SMB3 will run toward left or right.
In SMB1 and SMB2J,you start to push A button to jump and when you stop pushing A button even if you stop pushing 1 frame,Mario won't rise up even if you push A button again.On the other hand, in SMB2 and SMB3, if you stop pushing A button for a moment,Mario will rise up when you push A button again.
Tricks you can see in the movie
Super Mario Bros & Super Mario Bros 2J
Flag-don't-operate trick
Through the wall
Broken graphic in the water(pushes left and right on the same time.)
Vine through(5-2 in SMB2J)
Super Mario Bros 2
Bringing vegetables to boss room.
Vine glitch
Air jump
Super Mario Bros 3
Through the wall
enjoy :)
Flygon: Replaced YouTube module with higher quality one.
Baxter: A very well-done quadrun of the NES Mario games. Great audience reponse; accepting.
Amazing run! I like how the most attention was given to SMB3 since it was the longest, but three of the games finished at about the same time.
boct1584 wrote:
the only times there have been multiple Metroid games on one console was with Fusion and Zero Mission on the GBA, Prime and Prime 2 on the Gamecube, and (arguably) Prime 3 and Prime Trilogy on the Wii.
Don't forget Other M. Also, I don't suppose this would be possible for Kirby's Dream Land 1 and 2, but maybe Nightmare in Dreamland and Amazing Mirror, or Squeak Squad and Super Star Ultra.
You could also try Metroid 1 + Metroid 2 for a cross-platform TAS. The NES and Gameboy share control schemes so it's not like there's room for ambiguity. I think that'd work better than trying for more recent titles.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Holy cow, indeed it is great. Had to watch it twice for the awesomness. Voting yes of course and waiting for someone to do this but with the Metal Slug series.
You could also try Metroid 1 + Metroid 2 for a cross-platform TAS. The NES and Gameboy share control schemes so it's not like there's room for ambiguity. I think that'd work better than trying for more recent titles.
Metroid would be a LOT more difficult. Games like the MegaMan series and the SMB series involve mostly moving to the right with some jumping, so a lot of the work appears to be making sure all of the jumps either happen at the same time, or happen while the other game isn't accepting input (like between levels, going down a pipe, etc). The early Castlevania games would probably be workable this way too, although I'm not brave enough to try it :)
With the Metroid series, you would also have to make sure you were the player in each game was moving the same way at the same time. So if you needed to go four screens left in Metroid 1 and 7 screens left in Metroid 2 at a given point, you'd somehow have to delay Metroid 1 by the time it took to go the extra three screens in Metroid 2. Which I suppose is possible, but it would slow down the run drastically.
That said, this Mario movie is the best multi-game run I've seen yet! I was pretty blown away by the MegaMan 3-4-5-6 run, but that isn't so shocking because those are all pretty much the same game...these are different games with some character overlap!
Joined: 6/23/2009
Posts: 2224
Location: Georgia, USA
Zavalix wrote:
Holy cow, indeed it is great. Had to watch it twice for the awesomness. Voting yes of course and waiting for someone to do this but with the Metal Slug series.
This is the best idea I've heard all day :-)
Used to be a frequent submissions commenter. My new computer has had some issues running emulators, so I've been here more sporadically.
Still haven't gotten around to actually TASing yet... I was going to improve Kid Dracula for GB. It seems I was beaten to it, though, with a recent awesome run by Hetfield90 and StarvinStruthers. (http://tasvideos.org/2928M.html.)
Thanks to goofydylan8 for running Gargoyle's Quest 2 because I mentioned the game! (http://tasvideos.org/2001M.html)
Thanks to feos and MESHUGGAH for taking up runs of Duck Tales 2 because of my old signature!
Thanks also to Samsara for finishing a Treasure Master run. From the submission comments:
Shoutouts and thanks to mklip2001 for arguably being the nicest and most supportive person on the forums.
Voting no, for several reasons, but one big one in particular: the Super Mario Bros. part. Compare to the other three games, which finish at roughly the same time. SMB1 finishes way before the rest of them. Why not figure a way to extend the gameplay there (probably by taking less optimal warps), so that it can finish at around the same time as well? The same is done in SMB2, after all. That's the most notable problem I have with this run, and should be reason for reworking in itself.
A second, less important point: it may be an unpopular opinion, but honestly, multi-game runs are an obsession that need to be suppressed. The concept is neat, but the novelty is starting to wear thin. While they may be impressive in a number of cases (even if lag frame abuse is pretty much key), in these games there are different methods of moving between the games. I believe that's a first (X/X2 control more or less the same except for upgrades/weapons, MM3-6 are the same except 3 lacks the charge, FF5/6 are just walking on maps and menus so are pretty much the same). Here, SMB and LL control the same, but SMB3 is not quite the same, and SMB2 is pretty different.
The only hook is that they're of the same series - the Mario series. And with the common B-to-run and A-to-jump scheme a number of games have, you could theoretically beat like twenty games with one input. Where do we draw the line, though? How different would the games have to be before it's considered unimpressive? Or is the brand-name of Mario acting as the deciding factor here? Just something to think about, that's all.
System Error: you are of course entitled to your opinion, but I can't help but read your reasoning (SMB1's early finish aside) as roughly equating to "We already have umpteen million run&gun games; therefore voting no on this Metal Slug submission because it isn't novel enough."
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Joined: 11/17/2005
Posts: 278
Location: Massachusetts, USA
Is there enough time to beat SMB1 twice? The changes of the second "loop" around probably won't affect the TAS. But it would be pretty badass to claim that this movie beats 4 different Mario games a total of 5 times in 10 minutes.
So this is what broke the site last night. Obvious yes.
It is too bad that SMB1 finishes so much more quickly. Did you consider either warping to 2 or 3 in the first warp or warping to 6 or 7 in the second warp?
System Error: you are of course entitled to your opinion, but I can't help but read your reasoning (SMB1's early finish aside) as roughly equating to "We already have umpteen million run&gun games; therefore voting no on this Metal Slug submission because it isn't novel enough."
Well, that's a bit of a stretch. I was really just referring to multi-game submissions. By the way, given your specific example, you can never have enough Metal Slug. :)
Catastrophe wrote:
Is there enough time to beat SMB1 twice? The changes of the second "loop" around probably won't affect the TAS. But it would be pretty badass to claim that this movie beats 4 different Mario games a total of 5 times in 10 minutes.
I'd be all right with this. Even if the run does end up taking a little longer or being slightly more convoluted from the additional SMB1 input, in my eyes it feels like an even bigger cop-out to make SMB1 finish way before the rest. Whether the second loop or not using the fastest warps is the method used to extend it, though, it'd be a sure-fire yes in my eyes.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
That would go against the goal of "beat all 4 games with the same input in the shortest time". That would be a different goal: "beat all 4 games with the same input with an arbitrary time so that the endings synchronize"; one which can't be repeated or improved without great debate. The current goal is best.
For the record, I voted "yes". This is wonderful. I know the Mario games extremely well, and I don't know any of the Mega Man games, so this was much more satisfying than any of the previous identical-input runs for me. I am in awe of this author's brilliance.
Easy yes vote from me. I love seeing these multi-game runs. It really shows some creativity and knowledge of the games involved to pull something like this off. I hope to see more of these in the future.
As for the "SMB1 finishes early" argument, I have no problem with it. MM3-6 didn't end anywhere near the same time, so why should this run be criticized for being any different?
That would go against the goal of "beat all 4 games with the same input in the shortest time". That would be a different goal: "beat all 4 games with the same input with an arbitrary time so that the endings synchronize"; one which can't be repeated or improved without great debate.
There's a category for that though. ;)
"Contains speed/entertainment tradeoffs"
In other words, beating SMB1 twice would be an entertainment tradeoff for the speed. Plus it would be interesting how different all 4 games are when SMB1's inputs need to be put in after it's completed the first time.
I'm giving this a yes vote though since it was definitely entertaining.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
Axel Ryman wrote:
Lex wrote:
That would go against the goal of "beat all 4 games with the same input in the shortest time". That would be a different goal: "beat all 4 games with the same input with an arbitrary time so that the endings synchronize"; one which can't be repeated or improved without great debate.
There's a category for that though. ;)
"Contains speed/entertainment tradeoffs"
In other words, beating SMB1 twice would be an entertainment tradeoff for the speed. Plus it would be interesting how different all 4 games are when SMB1's inputs need to be put in after it's completed the first time.
I'm giving this a yes vote though since it was definitely entertaining.
If only that was actually a category instead of just a tag, I would be okay with it. I'd rather not have movies that do that obsolete pure TAS. I prefer the pure versions (yes, even for entertainment), personally.
This is awesome!
I never really like the idea of doing multiple-runs at the same time because it's difficult to focus on which one to watch.
But I guess it can be fun to watch sometimes.
This run is fun to watch.
Voting yes!
Joined: 4/2/2009
Posts: 376
Location: Porto Alegre - Brazil
Is someone willing to test this on the 'TAS in real console' device, with four TVs and four NESes? =P
Already voted Yes, .. but I too think it lacks a little SMB1 after the half, but.... is there a SMB1 10 minute route to be tried? xP
I was a bit bothered by the fact that SMB1 takes optimal warps and finishes 1/3rd of the way through the movie, but SMB2 avoids optimal shortcuts in order to finish at the same time as SMB2J and SMB3. Either SMB1 should take longer and avoid warps to end the same time as the other three games, or SMB2U should take optimal shortcuts and finish as quickly as possible.
The current solution is not consistent in its goals, which is one of the most important things for a TAS to avoid.
(My personal preference would be that SMB1 be made to take longer, by the way)
How fleeting are all human passions compared with the massive continuity of ducks.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
Upthorn: The author realized that only the longest game (SMB3) matters for the run's time. This creates an environment in which the other two games must sacrifice optimal time to allow SMB3 to finish earlier. Do note that this run is a mere 14 seconds slower than the current fastest SMB3 TAS, despite playing 3 other games simultaneously. The optimization for SMB3 here is extremely impressive. The other 2 games just happen to synchronize with SMB3 because of this optimization. That synchronization wasn't necessarily a design choice.
Actually, after looking at the difference in time between the quad run and the current SMB3 record, I became even more impressed with this run. What an amazing feat. Agwawaf is an absolute genius.
Edit: Wow. Compare that 14 second difference to the 5-minute-or-more difference all the other multi-game runs on TASvideos share. The MM3/4/5/6 run is 5 minutes longer than the longest game of the four. Agwawaf saw what resources he had for optimization and used them incredibly well.
If you got four NESes and four NESbots, could this be played back on real hardware? :D
Actually just one is required; the same input could be replayed on all four games in succession. Assuming the nesbot has the led input-showing feature, playing the videos simultaneously would still show that the same input is used for all four.
