Posts for Comicalflop


Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
haven't seen MM 100% yet (that probably needs to change), but OoT is way more stable... MM any% has so many desynchs for 2.5 hours, imagine a 5 hour MM movie?!? (if only someone would try and implement a dropped frame alerter like I suggested...) many people I know surprisingly do have lots of patience. no one has (to my knowledge) done a 6+ hour TAS yet, but I know of many members who have the determination to finish such a project. whether they want to sign up for it is another thing. knowing Zelda would help in optimization, but I'm under the philosophy that to a certain degree anybody can learn to TAS any game. Guanobowl had to start somewhere right? and there's scads of videos, YE's segmented runs to watch for knowing where to go, and GB's TAS, where you can watch input. can't be too hard right? it's taken the SDA peeps a couple of months to create the solid segmented route.... it'd be very tough to make TAS adjustments to what can be done quicker. there's boundary violations that can save a lot of time and change where you need to go next, uber tricks that weren't attempted that a TAS could pull off, and basic route stuff that's more suited to a TAS. (hard to describe, but there was a lot of routes discussed that a TAS could accomplish, that were dropped in lieu of a safer strategy/route.)
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
4 month bump, but I just started watching Yautja Elder's segmented vids, and read through about half of the 98 page thread at SDA... and have become convinced that a 100% TAS really should be done and makes all-medallion worthless. my current reasons why I would want to see a 100% TAS: -it inherently includes all medallions. albeit the temples will be broken, but you'll still get the joy of watching them TAS'd, and the bosses beaten. The temples may be done fast, and many rooms skipped, but you'll still see them. -all medallions does seem like "beat the game as quickly as possible, but for nostalgia take extra time to beat these temples, but make sure not to break them too bad." -it includes all dungeons, so Ice Cavern, beneath the well, and other stuff that is easily skippable will have to be done for heart pieces or skulltulas. so you'll see more of the skipped dungeons, besides just the skipped temples. also, you'll have to go through, beat mini bosses, and collect everything, rather than "run in, collect the nescessary any% item, then warp out." -Practically every part of the game will be explored/visited. This is appealing to me, and others as well I believe. who actually remembers exploring the waterfall in the Gerodu bridge canyon, and knowing it leads directly to Hylia Lake? the any% skips so much of the game, it's nice to have a run that goes to every location. (which is a general attribute of 100% runs in general, moreso when the any% counterpart skips so much.) -the goals put glitch restrictions, but in a good way. This is tough to describe, but "collect all masks/heart pieces/skulltulas" means that glitches to skip massive amounts of stuff aren't used (bottle adventure, beat only 1 or 2 temples etc.), but other glitches used to accomplish these goals in a faster time, many of which haven't been seen yet, which are very enjoyable. speeding up the gameplay/greatly reducing the time needed to collect all this stuff is better than simply skiping all of it. -I've looked at the route at SDA, and it is incredibly different from the any%. so someone having watched the any% will watch a 100% and see a whole lot of new stuff. -the cinimatic/Plotline to action ratio is way better than the any%. as soon as you get though the obligatory intro scenes, you'll actually separate from the plot somewhat in that more tme is spent doing non-plot things that ratio wise there'll be more Link action than blabbity blah scenes. -the "where the hell is the runner going" factor is kept from the any%; Link travels around Hyrule in a completely different random manner, showing again how possible it is in OoT to do things out of order. -the techniques in collecting the heart pieces/skulltulas/masks lead to new ways of using Link. with savestates/frame advance, these items can be collected in fast ways never before thought possible. -oh yeah, YE's segmented 100% run kicks major ass, and tool assistance can only make it better. -non assisted is aiming for sub 7 hours, I think a TAS can get low 6 hours or even sub 6.
AKA wrote:
Not me count me out.
in the SDA thread AKA mentioned how the most difficult part about a run like this is the route (besides just the long length). 48 pages of reading through showed me how many times the route was revised, and to my knowledge is still being updated every now and then. skulltulas can be day/night dependent; the biggoron quest requires a lot of sidetracking, and there's plenty of skulltulas/heart pices along the way; many HP/skulltulas also are spaced apart enough that figuring out which order of them to do when is tricky, sometimes YE goes back to get one later in a spot he already visited. sometimes something just plain doesn't work in a fluid manner (or at all) and has to be reconsidered. plus it may be that one thing that works in one place causes another event to not happen correctly in another.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
well there really is nothing that physically prevents you from doing so. if you have a big desire to TAS this game and feel like there will be enough of a difference, that's reason enough to do so. more banjo can't hurt, right? I was going to mention Banjo Tooie as an alternate, but the whole "can't complete the game in mupen" thing kinda reared it's ugly ass....
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
nitsuja wrote:
and I think it forces you to get pretty close to 100% just to beat it anyway.
I think this is in the same boat as Conker's Bad Fur day, where 100% makes more sense because any% is just too similar and you have to collect most of everything.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
the ninjas never stopped moving, bosses were annihilated like they weren't even there... yes vote (the first NES submission that I've voted on ever i think?)
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
My first TAS attempt was published, but I had to restart the whole run 7 times.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
"Do I have to go back there and push?" um, interesting idea, but the mechanics of being behind another CPU are very slippery. for one, being behind a kart is called "drafting" which will interestingly give me a speed boost and throw some things off, but more importantly, I don't even know if it's possible to push a kart forward. Ramming/moving into another kart, even at a slower speed, will push you either left or right (or up if jumping into it). there's a lot of bounciness when it comes to kart's coming in contact with each other. I made a test .m64 to show why it wouldn't work so well. I first used Bowser, and that was a bad idea, so I switched 2nd player to Yoshi. I think it's an interesitng idea, but won't work as well as F-Zero in getting a CPU to reach item boxes much quicker to make it worth it.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I decided to re encode the WIP using H264.... why I ever used xVid exclusively I'll never know. It's good to use superior codecs. don't know how it is on your end juggling 4 projects, but here's a few thoughts I've had since last posting: the 100% goal should be removed, because clearly, that is not going to happen, the monster density is just too much in future levels, even for TASing. The good part about mupen is that you can change the description, and change it back if 100% kills is possible. Definitely don't shoot the dragon's head first. Get all of his body segments, then shoot the head. Do reflect back boss projectiles. more damage, and it just looks really cool. and if a reflected projectile does hit a passing minion, so much the better. In very slow encoding mode I noticed more than just 1 tree that was missed... I'm pretty sure once or twice you even made a bunch of shots against a tree, then passed by it. For others, you ran right by them. For the trees you can't swipe with the sword, be sure to shoot them before you get to them.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
me neither. I mean, there aren't many s/he's that I know of on these forums.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I tried this in MMF. Here's what I concluded from a "not be in 1st place" strat: the item to get for such a technique is star. star has fastest power, lasts almost as long as an orange muhsroom, and still receives the same "faster when powersliding". But is dropping to 2nd/3rd worth it? My tests said no. It is simply too much of a sacrifice to gain a constant speed boost. when you're in 1st and not stopping for anything, you gain a surprising amount of ground. In LR I was vastly ahead of the CPUs, and the ones I encountered had actually all been attacked by other CPUs, which accounted for me lapping them so easily. In other courses where I'm not more than 1 lap ahead of the CPUs, they drive faster, but are still lagging behind me. as I've learned, in gaining distances/killer times, every frame counts. At the first set of item boxes, I had to wait 5 or more seconds for the CPUs to catch up and get ahead of me. that's a lot of time lost, especially considering that I can shroom at that point, and was almost at the next shroom. by the 2nd item boxes the CPUs were 4 seconds behind me because they had reached their max speed; again, still not worth it. Stars are fantastic for pulling off some shortcuts (KD lap glitch) and for going through obstacles, but it's best strength, for speed, is in vs. matches or in GP when you are behind. stars can lend to very amazing recoveries from behind. but it wastes time to willingly slow down at the beginning, lose your distance in 1st, just to get a star. However, I might get star in this run for other purposes. in KD, I'm going to get it to activate a lap glitch, and in KTB it's slightly possible that I can trick the game into thinking I'm in 8th place, and manipulate an item then. and there might be another scenario where I can use it to my advantage. a CPU getting star is actually a godsend to me, especially if a heavyweight and if the conditions are right like in the mega jump just posted. In the end, with extreme luck I may get it so that one CPU has a very fast start, and it is not as much time lost to drop in 2nd place and get a star, in which case I will consider doing that. but CPUs do seem to have slow starts, and Toad has a very fast start with a boost right at the beginning, and ssMT's/MT's, and usually reaches the item boxes within 3-7 seconds of the start. I don't think there is any course where I need multiple shrooms to pull off a shortcut. 1 shroom per seems just right. I'd also like to point out, for those who aren't aware of recent Mario Kart 64 development, that there is a very major shortcut for Kalamari Desert in the works. a shortcut this big/new is particularly interesting, since most of the major shortcuts for all courses that have shortcuts were discovered years ago, so this recent activity is actually very exciting, and I feel very honored to be a major part of it. I'm trying to introduce mupen to the testers as a good means of discovering new shortcuts, using savestates to easily pull off tricky driving, and easily capture an AVI, so it's a nice way of spreading the TASing love, even if not for making frame perfect movies but for test runs/demos. I also happened to encode a very seductive video from Weatherton, that cleverly uses the blue shell to gain a massive speed boost in the opposite desired direction. KTB blue shell boost This is a very sexy boost, and if only the shell had hit Toad on the right side, would've very likely made it to the other side of the mountain. KTB is very complicated. There's 6 or more shortcuts/glitches that I know of, many untested, many having not even been successfully completed, so there isn't even a route because it's unknown exactly which shortcuts/glitches to use, and in which order/lap to do in sequence. Figures that the mushroom cup ends up being the trickiest with so many shortcut possibilities and tech.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
*slip ahhhh!* ouch... guys, it's not nice to leave puddles of testosterone lying on the floor like that, someone could get hurt. let us know when you're doing being children having macho contests on the playground.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
hrm, something's not quite right... it seems to have desynched, and become better optimized as a result gaining 101 frames. how interesting.... Is it me, or does every time you post a WIP I go and find another way to abuse the ninja? just when you thought I was done exhausting all the ninja moves, Here's some more! I sort of realized that all these can only work if opponents/enemies can be damaged every other frame...
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Loved it for a few reasons: you deserve, after the painful months and months of facility, to improve the times on the missions you've already done your progress is getting faster, which I bet must feel good (for both you and us) and lastly, it was very optimal, and I actually got to see more of the level this time around. Bond shouldn't be shy about staring his opponents in the eyes as he shoots them in the head. The look down made it seem as if he was mostly concerned if his shoe was untied or not.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
since JFG is stalled and I won't be touching it for a very long time, I thought I'd at least show everyone the Eschebone sequence break. Crisper .WMV .m64 (fast forward to Eschebone)
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
so THIS is what older PC games were like. huh. (yeah, I'm in the "haven't seen this game" category. or heck, "didn't play DOS" category as well.) a new system for TASing is exciting. This game to get the ball rolling not so much. Voting meh because I have seen wayyyyy worse movies; the many doors it can open, plus the fact that it's pretty optimized/excellent route, and those nice space battles makes it interesting in that regard, but not worthy of a yes.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
One of the many reasons why I hate Mario Kart TASing of any kind: not as bad as the only item you want appearing on a specific in-game time, but I need to hit the item box and be driving in the exact right way for a mushroom to come fast. what may be perfect could end up in a very long wait for it to scroll through and select a mushroom, at which point I'm already at a steep hill. blarg. hrm, those are interesting ideas... but again, getting the right circumstances for a mega huge jump like that seem astronomical. if it's the CPU being under star that makes it happen, and not necessarily a boost effect, then it will be even harder. and Weatherotn and I took the 540 theory but couldn't apply it the way we wanted, and maybe this same theory won't be able to be applied the way we wanted but still used regardless. but phew man, what a breathtaking jump. that has to be worked into this TAS somehow, talk about truly surprising and not expected of this game.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
well, I do have an idea or two for Royal Raceway, but the odds of my desired CPU having a star so early into the race are nil. I've so far manipulated a CPU to not use lightning when he/she normally would, and have made someone's star disappear. forcing a CPU to have star seems maybe a little out of reach. I can't divulge much more, but I'm slightly certain I can do the 2P shortcut at least once or twice in RRa. I'll see when I get there.
Spider-Waffle wrote:
I wonder if you could launch around the finish line this way.
we tried that with the 540's vid (which unfortunately we couldn't do, Wario was juuuuust barely not close enough to the side of the road.) It would only work if the extended invisible finish line is incredibly short on one side, which is only the case for 3 tracks: Luigi Raceway, Mario Raceway, and Wario Stadium. MMF update: perfected the first 9"86 of lap 1. Now I have to test to go up a hill or around it, and which is faster.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
well, after having watched some of it, I like where you are going with this. However, as with many N64 racing games this has a *lot* of room for perfection. I'll get Fried to look at this, but be expected to have to get as close to perfection as you can. also, would it be better to not have raw input enabled?
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
some will not care, some will care too much. in the same vein that the stars for recommended movies weren't intentionally made for people to fight over (and to this day it still isn't meant for competition), and now is something that some (not all) people strive too much for, some people will try to hard to become an expert player. the competition should be in the obsoletion of movies, not in rank/status. in a forum with many people, the reactions to such a change will be varied.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Kyrsimys wrote:
Maybe the rank should just say for example "Player, 6 movie published". Of course the system would have to be automatic, it would be too much work to update the ranks manually.
There already is a separate page of sorts that already features this, namely the players page.
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I get the same. Let me try right settings, like Pilotwings this game desynching doesn't feel right. Edit: Jabo's 1.6, still desynch same spot. when making this, are you replaying the entire .m64 from the beginning to check for desynchs?
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
wait wait wait, DKR progress and I didn't notice? that's got to be fixed! will take a look at these right way!
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
silver bullet it isn't, but it would be a much better way. knowing a desynch just occurred during recording is much more helpful than you might think. As Mukki will tell you, he'll do some progress, maybe a minute's worth, usually less, then have to play back the long movie in 1000% speed, and check for desynchs. This obviously wouldn't be a permanent solution, but a very helpful one to have, especially since likely you'll be doing frame advance, and you'll be right on the spot, during recording, of where the dropped frame is and can immediately fix it without having to replay the movie and lose any progress. can such a program be made or feature installed?
Homepage ☣ Retired
Post subject: N64 desynch Tool: Dropped Frame Alerter
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
It's been suggested in IRC that one possible way to avoid desynchs is not necessarily stronger software that can handle desynchy games, but in knowing when a frame is dropped, which seems to be the case for why N64 games desynch. (This has been shown to happen in both Exp Pak and non-Exp Pak games.) This would be a good, early solution before a better emulator was made to handle fragile games. If there was some way to detect when a frame is dropped during recording, and possibly a visual alert was made saying "Frame #12728 was dropped" in bright red letters or something, that might help a TASer in knowing immediately if a desynch occurs, rather than have to play back the entire movie in read only mode and discover that a larger portion of work has been made useless due to a desynch earlier on. an alert during recording would mean that you'd instantly know if a frame was dropped, you would still have a savestate before then, and already know the frame where it occured, so you can immediately proceed with caution. This would save a lot of time and effort on the TASers part. Could a feature built into mupen, or an external reading program (like Gocha's external mem watcher) be used to monitor dropped frames?
Homepage ☣ Retired
Editor, Experienced Forum User, Published Author, Experienced player (730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I was more intending my site to be for the members here to view, since most don't know about the wrvids.com page. (I did, it's how I got access to faster vids than at SDA for MK64 to study.) It will also not have as many folders to have to sort through, and most importantly will contain entire laps of individual courses for the runs that are in progress, specifically to address those people that may have a hard time watching the progress. On a related note, I have all 10 of Huffer's completed courses encoded, and working to get them uploaded to youtube so that I can module them onto my site. To answer waffle's question, an infi-boost cannot be done right after an extended boost, unfortunately. you need to release a normal mini boost into a speed strip of some sort, and then not jump or bump into anything. as long as you do this, you'll have the mini boost's top speed, and not drop down to normal top speed.
Homepage ☣ Retired