It's difficult to make a worthwhile TAS of a rhythm game, but this challenge is fairly well-suited for a TAS! The scoring system behind the DS game has always been shrouded in myths and misconceptions, so it's interesting to see exactly how the game can be contorted when one knows exactly how it works.
From my understanding you didn't need to do any "early/late"s, but you also chose to go for the lowest possible score, right? Was it to save time? This way there are no scoring comments at the end of each stage, and Superbs are bad because they create Perfect opportunities later on which slow down the menuing. Or did you just do it for entertainment? I gotta say it's downright hilarious seeing stuff like Crop Stomp being accepted with 0 crops collected. Or is there a technical reason with the scoring that I didn't catch? This is something I feel should be elaborated in the submission details.
What is the purpose of the RNG manipulation? I only ever knew this game as being random in one spot, which is the order of the ditties in Big Rock Finish. Is the "invulnerability period" you get after a miss random?
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
The extra HUD encode was surprisingly entertaining. It's pretty elaborate, and really shows how complex of a game Ecco 2 really is. I'd love to have a more in-depth explanation of what all of its elements are.
As for the run proper, it's just as great as the previous one! Easy yes vote.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Well, I'm happy to see all levels being tackled (which definitely means this run should be in an "all levels" category), but I'm still irked that the very first world is skipped like that.
Normally, the idea of a category where the runner goes everywhere is so that the viewer can enjoy as much of the game's content as reasonably possible, right? Skipping an entire world but then playing all others normally, even if it's a legit mechanic in the game, just feels like it defeats the purpose of an "all levels" run.
In terms of entertainment and execution, much like the previous attempt, this run is fantastic. Some of the new time-saves are also so juicy I'm still salivating. I'm just not too pleased with that specific world, personally.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Sorry I wasn't able to respond sooner. Yep, I remember that when I was a kid, I discovered at least three doors that could be skipped, but I wouldn't put it past this game that there could be more.
To make the definitions of "any%" and "100%" clearer, I ask this: Would it be possible to have a run that goes through all acts? (I'm calling an act any location you enter when you touch a hub-world door or barber pole exit, and the screen fades out/in.)
If so, an any% TAS could just be about getting to the end of the game ASAP, whereas a 100% run would require going through all acts; this implies that the runner cannot just take the left barber pole in world 1, and would also cover "entering all worlds" as a requirement.
Whichever way you slice it, I'd say this particular submission is neither any% nor 100%, despite being a pretty good TAS. So I too think that it would be best to cancel it, and go for a new one with clearer-defined goals, and with better level-skipping knowledge.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
I could swear I had brought up this question before, but a search proved fruitless:
Since this is an any% run, as opposed to 100%, shouldn't the optional zones be skipped? Some doors do not need to be entered in order for the next doors to unlock.
Or would skipping them count as deliberate cheating? It's not much different from what's done in the first zone anyway, when a runner uses the "secret" exit to finish it in a matter of seconds.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
You'd probably be disappointed in a score attack. Most of it entails just bouncing against ceilings to abuse the ribbon mechanics.
I see. There's always the possibility of somebody creating a TAS that doesn't abuse this exploit. Though such a specific restriction might not be accepted at TASVideos, at least the resulting run could still be interesting, provided there are no other broken strategies.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
I too would love to see a score attack run one day. But for a speedrun, this accomplishes the goal of being fast while seeming fairly polished, using some tricks and glitches that are sure to entertain most players, and overall being fairly clean for an otherwise more chaotic game. Yes vote, easily.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
The time saves are clear, the ways to avoid collecting stars are very creative and fun, and the time at which this run was published makes it possible for fans of the game to compare the SNES run to the Mega Drive one, what with the SNES run still being fresh in people's minds. Yes vote for sure.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
An Emergence Cave run? TRex Quisite did it, in 13 days, apparently:
Link to video
I made some calculations some years ago, and it looks like it'd take over 20 hours to do this. Yeah...not fun.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Probably the most important bit: I can tell this isn't an all level run.
Oh no, I wasn't saying that the fact that it isn't an all-level run is bad. I said that to tie in with what I say next. Basically, it's me pointing out that I DO understand that it's an any% run... But I just don't understand why you don't skip the levels, if it's really an any% run. That's the biggest gripe I have with the run. Like I said, it's good stuff, but this should either be an any% run that skips all optional levels, or a 100% run that does all of them. Not a sort-of arbitrary in-between.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Well, I can see optimizations in it, but also lack thereof.
Probably the most important bit: I can tell this isn't an all level run because you don't say so in the submission, and because you don't enter some secret levels. So, you should've just skipped levels 1, 4, and whatever the one with the tabletop games is, by not entering their doors.
You should try to reach the end of the level with only one peg of health left. That'll give you a smaller energy bonus at the end. ...At least that's what I think that is. Try something similar for the mystery bonus, too.
On the level with loads of blue metal girders, you go on top of some bell-like blocks on the ground that are connected to long vertical chains. Does walking on these make you lose speed? You should memory check if you're not already. Check your speed in other places too.
Finally, it's somewhat lacking in the entertainment department for the first half or so. In stretches where you just walk forward, you could break the monotony a bit by jumping or something. On the autoscrollers, you could try standing on the obstacles until the very last frame before you get crushed, react to stuff before the audience can even see it on-screen, circle around a collectible... the entertainment catches up later on, but you have the tools to do super-human things, so do them!
That said, you had a few brilliant moments, like the transition between the Christmas level and the music level.
Sorry, I'll have to vote no because of what I said above. But I can't wait to see your improvements! I want to see a really good run of this game, and you're almost there.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Yesterday night I found something interesting by messing up with the spot
http://imgur.com/a/eAwDv
I definitely didn't know lag frames could've been graphically (and not only I think) different from the previous input frame. Shoutouts to Biz
Looks to me like it swaps and draws the buffers every frame, but doesn't update them if it's a lag frame. So it's just temporarily drawing the previous frame.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
I stumbled upon a bad ending TAS yesterday, with the aim to get 0 parts.
Link to video
It's not very optimized, and I'm pretty sure the tricks used are old, but still. Just remember to fast forward at the 2nd day.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
erm I personally wouldn't trust anything on that wiki if I were you, there have been quite a few things wrong. Take everything there with a grain of salt (i.e. white pikmin do not attack faster <_< )
What? That's ridiculous! Where did you read that? Removed it.
You know, just because one thing is wrong doesn't mean the entire wiki is crap. This wiki was built over the course of nine years. The information within is accurate. Of course the occasional slip-up exists; remember: wikis are places anybody can edit.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
But on the Wii version, in addition to the cursor being disconnected from your movement, you can also skip the end-of-day cutscenes, saving around 34 seconds per day (a little under 3 minutes over a 6-day run)
Actually, you can skip the end-of-day cutscenes in the European version of the GameCube game. Over the last few months, I tried documenting all I could about the differences between all versions of the game, both region and console-wise, as well as some other technical tidbits of the sort. I put them up on the wiki, so these pages will be very helpful (even if sometimes incomplete):
We also have pages about glitches and such on the wiki. Like PiePusher said, TASing Pikmin is NOT easy. You need to know every little thing about the game, if you want an optimized run. I'm no speedrunner, but I know the ins-and-outs of the game considerably well, so I'll try my best to help!
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Oh! I knew there had to be a reason to miss the shortcut!
Anyway, that ostrich is really buggy, isn't it? First, it gets two necks when the ride is over, now this. Wasn't there a third glitch too?
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
I'm going with a Yes vote. It's quite the run, and improves upon the previous one quite a bit. The out-of-bounds water current trick made me go "woah!", not particularly because of the trick per se, but because of the sheer amount of force it took to do that! I knew those currents were strong, but that was just insane!
Judging from the stream Patashu linked to, quite a few tricks done on this TAS are pretty similar to the camera freeze glitch. Which makes me think, would the goal to not use the camera freeze really be that important? I mean, some of the glitches used here are almost the same thing, and there are other glitches throughout the run that are repeated often (crystal sliding), which rules out the "it becomes repetitive" argument. Plus, games like Sonic 3 & Knuckles have TAS that use wall warps in almost all stages, and those are still enjoyable.
I can't imagine there being such a monumental difference between this run and a camera freeze run. Hence, I'm not really comfortable with the idea of this run avoiding it, and the potential for another run to use it.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Normally, only people who know a game actually watch runs of it. It makes no sense for me to watch an Okami run when I don't appreciate the tricks, skips, or what makes X faster than Y.
So if you already completed TWW, and know it well enough to appreciate the TAS... chances are you REALLY don't care about the plot, because you already know it. That, combined with a broken story sequence thanks to the skips makes following the plot on a TAS a ridiculous idea. So who cares!
Not like it matters for this case, because the JP version clearly has to be used; as Tompa points out, it saves 15 minutes. But I'm just throwing in my two cents for general "English vs Japanese" discussions.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
How would that work? "Minimum A presses for maximum amount of stars except skip it when it's too long"?
To me, it seems arbitrary to skip some stars, however way you do it.
Uh, crud, I didn't mean skip skip...
All right, because BLJ is out of the picture, 70 stars are necessary. So, the 70 shortest stars are picked, and the remaining 50 are "skipped". That's what I meant.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
Hmm, I see that. Though I still think it's a tradeoff -- you're trading speed for entertainment. That's what "tradeoff" means, unless I'm wrong.
But yeah, something as arbitrary as "minimum A presses except when it's too long"-run would be stupid.
I guess we just wait for faster methods...
But hey, what about skipping the longer stars altogether? That's still in the table, no?
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
The RNG DOES work with a seed, but you can still manipulate luck to a degree, right? Suppose you want to force a Manaphy to come out. The seed for the fight generates the numbers 335, 681, 999, 0, 123. You want to open the Poké Ball when the RNG is 0. Well, you can push the numbers forward by doing a couple of things, like using Judge, or to a more sophisticated degree, moving differently next to a computer player, so that it picks different random actions (a CPU next to you might randomly jump, run, etc., while a CPU next to you being attacked will randomly decide to get hit or shield).
Sure, it's next to impossible to push the RNG in such a small event but... Could the RNG be pushed with button inputs also? If so, one could try to mash buttons on the READY? message to manipulate the RNG...
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
But this is all for a 100% low-A run. Surely a 70-star low-A run can be made while skipping all of the insanely long stars, right? Off the top of my head, I can only think of the Bowser levels taking a while to set up, because of all the Goomba ladders.
Even still, there is such a thing as speed-entertainment tradeoff. For stages that you just cannot bypass, instead of taking 2 hours to set up enough steps for a jump, just do the jump already. It's less impressive, but it's a necessary tradeoff, so that the movie does not drag on (implying a non-skipping encode).
Of course, this can all be debated once the run actually takes off.
I'm Espyo from the SRB2 Forums.
Current project: A Pikmin fan engine, Pikifen
