Yeah... on console the BoB BLJ is very tricky to get going. When you succeed you usually end up abruptly warping, or don't get enough speed to make a wall. I'm wondering the easiest place to pull it off, since outside the castle is no picnic either. Not much use in a live speedrun at the moment.
I really suck at tasing this game. I just messed around with the rolling cubes outside the pyramid. That's some hacky collision code there. I was slipping all over the place, clipping in and out of the box and sometimes insta-dying in the air for no reason. No luck getting a BLJ though since the hole inside is bigger than I thought. And I guess if one of the signposts could be used it would be known already.
What about using the walk-walls of a pillar in conjunction with the bird? :/
Lol. I was going to suggest trying a BLJ inside the tox box but then I remembered how large they were as well. Looks like bobmario managed better anyway. I figured any successful SSL BLJ would require clever usage of multiple points of bouncing after the initial start. Unless there's some obscure place on that massive slope by the red coin behind the start where you dont slide.
I don't really know if you can call it double standards instead of just nasty irony.
You can't, since adelikat has admitted to have not noticed the post about improvement.
Besides, scaring HappyLee away on his 11th submission for SMB? …Right.
Go to SDA if you want to "delay" publications just because someone else found an "improvement" (that no one in the whole fucking world will notice anyway).
The fun of watching this thread get into a flamefest aside, this line here made me laugh. This never happens on SDA over 2 seconds of improvement. In fact, usually the suggestor gets told a few choice words about who he can go fuck.
I actually do agree with the rest of what you've said in this thread, though (although not with the tone of it). There's a thread in general right now about more or less the same sort of issues. I find it odd that this non-optimal improvement would get accepted in the face of new finds, but I don't really know if you can call it double standards instead of just nasty irony.
Certain parts get more interesting if you watch in something like media player classic and slow it down to 1/4 speed. The cursor movement is indeed optimal.
I actually did miss the fun of the bonus round the first time I watched. I now realize the fun irony in the first half of it that 5 totally plausible and/or true answers all scored 0.
I just about lost it when you started dropping ghetto language and comparing Family Feud to The Price Is Right. Bonus round was beautiful too. 9.5 For entertainment, not sure about technical quality. This movie is pretty out there but I think it needs to be on the site.
You are right. Was just trying to brainstorm the fastest way to complete all levels, but as you said, the warp glitch adds a level of ambiguity inconsistent with the goals.
Its a shame this wastes so much time that its impractical, but it'd go pretty smooth. Warping to 6-8 like in the glitch tas, beating it, then going back to world 1. By the time he gets to 6-8 again it'll be like an hour or two later so I could live with watching it a 2nd time. If it has to be 2 times nearly consecutive, eh fast forward's there for a reason.
In that vein...
Is it possible to use the warp glitch to warp to bonus stages before they're unlocked by getting 100% on a world? If so, 6-E could cleverly be performed right when world 6 was beginning, in the course of a 100% TAS. It might involve backtracking to a stage like Gusty for the required coins, but it would ultimately exit with the map on world 6. Net time loss ~20-25 seconds at most.
First off, let me mention that my personal choice of saying "tass" like "ass" with a T at the start comes from a local company called TAS electronics, which pronounces their name likewise in advertisements on TV, and has done so for years.
Now, for the meat. I completely understand how TASing is done. I know my way around an emulator, have experience with programming, and have used hex editing to varying degrees to analyze computer games, make trainers, or just give my gf infinite money when she's playing the Sims. I am totally capable of making a TAS, and I love watching them. TASing has me aspiring to attempt practically impossible things like 1 star SM64 on console. I come from the SDA community but appreciate both types of runs as seperate and different works.
But I just don't see the point of even trying. It seems impossible to just sit down and decide that with a relatively small amount of trial and error (100 -500 rerecords per second of finished product) I can find the absolute optimal route through a game. This especially sucks with 3d games. The other thing is that, particularly with games that have been worked on before or have existing movies, finding the fastest route or a clever improvement requires incredibly out-there and counterintuitive methods. I could look at the speed variable in SMW2:YI all day long and probably never figure out the proper left right presses to accelerate optimally... they would just never occur to me... why press left to go right? that sort of thing. Wario Land 1 is a game I've loved since I was a kid and greenalink is rapidly and incredibly both discovering glitches and pulling off tricks I would have never guessed in a year of work to attempt.
That and the type of standards + criteria for acceptance here makes the prospect of creating a TAS one of those "fuck it" kind of things. I'm more qualified than the average visitor to create a monster TAS and I have yet to read an encouraging argument, useful bit of tutorial, simple explaination of the basic process (as though I was about to sit down and start grinding it out, not just an overview), helpful encouragement and suggestions for a beginner... this place is not noob friendly. You can look on youtube for simple videos about how to get a sideways stair blj going in SM64 but at the end he doesn't go back and give pointers for how it could have been optimized to start in fewer frames. I know how to hex edit but I would imagine I'd still have to learn both the interface of the emu's particular editor, and what I'm looking for and how to find it in a console's RAM instead of a windows process.
Some might argue that the best way to learn would be to attempt to replicate a simple TAS of a solved game (throwing in a game/math theory term here... I think we should use it, even if it's improper, for a game that's generally agreed to have a 100% optimal TAS using all current known methods, strats, and glitches... and it remains in that community-limited solved state until a significant new method is found). I don't want to reinvent the wheel, though. I can watch a movie with the inputs on screen (I actually prefer this) and understand almost exactly what the author's doing and why he's doing it, but that doesn't bring me a single step closer to feeling I could begin my own... it only reaffirms my conviction that it would be a waste of my time to just copy his movie and see if i can arrive at the same efficiency on my own.
If I can get an answer or even just a little followup to all that, well, that's the tip of the 'iceberg', but it'd be a wonderful start.
Yeah JRB has become quite the level (about to watch the posted WIP, but I remember earlier snippets). DDD is now the most boring out of all the levels, I think... followed by rainbow ride for the repetition involved... and it doesn't help that the last two levels of the game have the same damn music, either.
Videos posted on the previous page were awesome as well. I like that new tricks for star times are continually being discovered. Sslayers's WF sideflip wallkick is one of the crazier multiple-use timesavers I've seen.
If it plays on a pentium 2 but not well on a late model athlon xp it's either some hideous software problem or a hardware issue with current generation video decoding on certain processors / certain motherboard / video card setups. I remember my XP 2400 + radeon 9800 pro, it hated decoding any kind of newer video (youtube, .264) without going into high CPU usage. But that's very odd. You'd think trying different codecs or players could alter things.
can't speak for others but my P4 plays SDA HQ at about 5-10% (but only encodes to sda HQ format at 1fps, 80% usage)
Here's another few offsets I discovered:
- 0xA91A: Wario's current state. These are written whenever Wario does something (eg., jumps, walks, dies). Here are the appropriate values for this offset:
- 00 do nothing
- 01 walk
- 02 crouch
- 03 ??
- 04 swim
- 05 jump
- 06 on a ladder
- 07 bounce off of a door (treasure/goal/gate in course 40)
- 08 jump?
- 09 die
- 0A don't know, changes to 05
- 0B bodyslam
- 0C bounce backwards
- 0D running jump
- 0E shifts a few pixels forwards
- 0F throw something, changes to 10
- 10 ?
- 11 get stuck, jump to get out (quicksand?)
- 12 get treasure
Any higher values will cause the game to reset or freeze.
- 0xA942: Scroll direction. 0 = scroll up, 1 = don't scroll yet, 2 = scroll down. This can be used to warp to odd glitchy places, but could also be used to get the screen to scroll.
Some of those unknown states are probably holding something, holding the hand up, jumping with hand up, ground slam, flying posture, dragon flames shooting, crouching, crawling, gloating at treasure/coins... etc. ... in fact, the fuel counter, or a slot near it, could also indicate how many frames of bodyslam or fire are left.
Heh, this is excellent. and don't forget the (small amount of) time you save not collecting those extra rupees to buy the bomb now.
But the tempered sword still seems like it's worth it, if it does in fact cost ~4000 frames (a bit over a minute). I'm assuming it does at least master*1.5 damage in making that claim.
I saw this vid on youtube, the people at the gamefaqs board found a way to kill ganon without the silver arrows. It is to basically hit him with the sword right before he starts to teleport away. If used, it could impact route choices since they big bomb could be skipped (and even getting your sword tempered, although that may still be worth it).
Hmm. Well, the tempered sword is worth its upgrade time for the extra damage it does, but skipping the golden sword (not sure if you get it anyway) and silver arrows would be nifty (assuming it saves time vs. killing ganon the normal way). Also means you wont need to collect any more rupees.
I was going to say something positive about the improvement, but then I read two posts in this thread that made me fall on the floor laughing. You guys really created a gem here...
Immortalized in a screenshot: http://img.photobucket.com/albums/v469/drybes/wow.gif
"I would go with this" ;)
I would get the first Joker and then the lower left Joker (2 questions out of the way of the path, for two more challenge skips), but the upper left Joker is unnecessary (3 questions off the path) as there is only one timed trial to skip after that point. Unless you can't manipulate what trial it is for some reason (perhaps the trials are determined by the frame you enter the furnace fun room, or perhaps they're determined when the entire game starts, like the set of grunty-specific question answers is). But yes, the flibbit (i think that's what they're called) trial takes like 10 seconds with wonderwing. Even without wonderwing, it's worlds faster than either flipping floor tiles game.
Hopefully something interesting happens when you try using a joker while in mid-air, or using a joker and jumping simultaneously (skip multiple squares?). The invisible walls around the entire board probably aren't going anywhere.
As unentertaining as a TAS of jeopardy would be (ever try playing it normally? yeah.), it would be cool to achieve the highest score possible in the game.
At least on the real show, a daily double is never placed in the lowest value row. I think it may be restricted to the 3rd row and down, but I could be wrong (it could also change between rounds).
But yeah, let's assume the 3rd row is the lowest dollar value it goes to. So all you have to do is manipulate it to one of those slots and then proceed to answer every other question first for the round.
Total score for round 1 is 100*6 + 200*6 + 300x5 + 400x6 + 500x6, then daily double takes score*2 = 17400.
Double jeopardy is the same... 17400 + 200*6 + 400*6 + 600*4 + 800*6 + 1000*6 = 34200*2*2 = 136800.
Final adds the additional double to 273,600 as the highest possible score on Jeopardy. Suck that, Ken Jennings.
But yeah, it's an open question whether scores above 99,999 will even display. The game probably would keep track of it and might show it on the game over screen, but the score counters from that era were only capable of 5 digits.
--
Considering that's the absolute most interesting you could make a game of jeopardy, I'd say don't waste your time unless you're doing it for personal reasons. And if you are going for a time attack, I think it would be much more amusing if you manipulated both computer players to answer wrong and end up in the hole by the end of double jeopardy, in which case you'd win by default and not have to sit through final.
What settings I have to set, lest BLJ-glitches works in this game (in Project64 or Mupen64)?
Erm... you don't need any special settings, the default ones work quite well. I wouldn't use Project because you'll need frame precision and probably rerecording to pull off any interesting BLJs, which is how that video (and all others here) was made. There exist tutorials on constructing basic BLJs in mupen if you search on youtube. Other than that, the only think you need is the rom, mupen64 rerecording version out of the box, and the TAS input plugin greatly simplifies things.
If you look at the history of TASvideos, almost all published runs that were thought to be more or less unimprovable were obsoleted by new runs that were several minutes faster in the end. This is especially for longer runs like this one the case. The difference is that the other authors just weren't aware of the improvements, so they obviously never stated them beforehand.
Hmm. I was pretty sure TASvideos didn't accept movies that, upon submission, had significant known improvements to multiple parts possible.
I'm sure that the people who invented the telegraph weren't aware that a hundred years down the road the telephone would be invented. But what you're doing here is standing next to a telephone, trying to sell us your old, battered telegraph equipment, even as you tell us it's obsolete, even as far as telegraph equipment goes!
So again... even assuming that no one has any problems about whether or not the category of this run is a good idea... it is not as fast as it could, and should, be.
What's the issue here? Arbitrary category or not (I think it's a legit category), by author's own admission there are improvements possible. Does that not mean then, necessarily, it must be rejected? It's not like the guanobowl OoT situation isn't the same thing. It's a fallacy according to the guidelines to accept a movie purely because it is an improvement...
It's a good movie, but I think you knew, even writing that description, what would happen... it shows in the writing style you assumed.
Certain enemies are more likely to drop 100-coin coins. I've gotten 3 or 4 to spawn on a single level before. I think it might have to do with the type of enemy as well as how much immediate peril wario is in, like position of other enemies near him. This is all from console playing so yeah... I dunno what else to tell you. For what it's worth, I have the regular GB and not the GBC version, so there might be a difference if they noticed the GB version was particularly lax about spawning such coins (although I doubt it). Other than in world 1 or the early, easy levels, I typically encounter about one per stage in a normal playthrough. What do you need coins for, anyway? You can guess the number correctly at the end of a level every single time in a TAS, and the in-stage minigame only costs 50 coins on Hard (which is also the fastest incarnation of the game of course). It's piss easy to get 50 coins on most stages before reaching the bonus.
Also, I would love a 100% run. Specifically, it could start on the regular 25 stage path, then do all the side stages in order. I also think use of the reset button or ABStartSelect or something lets you avoid all the mandatory cutscenes at the end of sidequests, because the game still saves the file to the cartridge memory... however you have to configure the emu to reproduce that effect. Definitely not a bad idea to work on a normal 25 stage run first, though, if only as practice. I love this game.
The highlight of a 100% run would be that final time attack stage, which is rather difficult to play in realtime.
I'm downloading the WIP right now. You are doing some very awesome work in a relatively small amount of time and I salute you :)
I'm 99% sure you know this already, but you can beat the genie as jet wario, because you jump high enough that a jet slam with B actually damages him, allowing you to do away with that silly lamp, so make sure you're Jet on that course, regardless of how suboptimal other parts may be.
I will probably be attempting the console version of the 100% run but I'm curious what strategies you will use so I'm waiting...
Hazy Maze Cave...falling onto the purple ! switch from a really high distance, perchance?
Has this been looked into any? It's fairly easy to get high enough so that your butt-stomp gets you far enough underwater to hit the switch before you start swimming, but thats just what I remember from the console, may or may not be relevant. Does it have anything to do with the physics we're trying to work with here, or is it just a case of hitting the switch because your downward momentum is fast enough?
I'm also curious about what causes vertical zipping behavior (i.e. teleporting to the top of the pole in BitFS in one frame), whether vertical seams can do this, etc, but I'm sure its been looked into.
Kirkq wrote:
Judging by the rest of the video, he took damage from the flame, which essentially places damage on you over time, and then died just before/while jumping into the water.
We could feasibly recreate that by the slope outside the castle by jumping into the water from high up with 1 health left and killing ourselves underwater, but I doubt this would be beneficial seeing as in the video he just died and respawned. Interesting video though.
How nice would it be if you could die normally underwater like this, in an area where you respawn in the same area (i.e. outside the castle), and respawn in a swimming animation?
But yeah, things like this should be proof enough that certain animation routines cause various other things to be overlooked. You can fall into a bottomless pit and continue having a flaming ass, long after the health depletes to 0, and the death animation doesnt really seem to care if you transition in and out of water, etc... there's probably a way to time touching the low water level switch in WDW so that you lose all your health, drowning, right before the water level is low enough so that you'd be standing (i.e. on a slightly higher place like by the blue coin switch), and I would imagine mario does the drowning-death animation on a dry surface. Nothing short of death and a burning ass, in my memory, have properties like this.
Whoever it was that mentioned getting Mario to take damage underwater, what exactly did you do to do this?
My only guess here would be that you'd have to fall from height, land on something, and slide off into the water while damage was still registering on the meter, which is probably not adequate. Unless there's a surface that registers hitting it as damage somewhere on the way down (such as inside that column that leads to the opposite side of the moat door, but then you'd be on the wrong side of the door)
mr_roberts_z wrote:
ThMrksman wrote:
Basically, either krohn7master will make a .m64 showing the glitch and someone will start making a new 0 star run
This won't happen. He has already claimed that the moat glitch was a once-in-a-lifetime thing, and there was little chance he would ever get it again. Meanwhile, even if he could prove the BLJ into BitDW with an m64, it would be far slower than what we have currently, and pointless in a 0 star TAS.
Yep, it's only of use to a console player, and from the looks of things, the pillar BLJ is still easier even if this turns out to be legit, since it's apparently not too easy to do...
I don't suppose there's a way to glitch yourself into the cannon/cannon hole, but do cannon physics allow more headroom when landing in water or are things the same? I was trying to think of some way to get completely below the landscape but now that I think about it this doesn't achieve anything of value, I've been thinking of the invisible water in OoT.... oops
I am no longer sure if he's a scam or a guy telling the truth. However the controller wasn't within the frame at all times like I requested. Could you do that for me krohn? I kinda wanna see what buttons you're pressing and what the back of the controller looks like. Appreciate it! If you could also press Reset or something similar at a random time, would be stellar.
I think at this point people of both opinions would make better progress to get off his back a little and wait for the next little bit. By stating he did it in Mupen he removed what would otherwise have been a legit thing to claim.. that one of the contacts on his cartridge is bad or soething. So we're really just waiting for the next video, and in the meantime hold your tongue, even if you know you're right, it will help.
