I was thinking the best way to do it would be to make one shallow pass to pick up something (any weapon) worth carrying into for the second, full run. Each run probably wouldn't need more than 6 or 7 passes to find something acceptable, although I agree that if you want something that's "ideal" you'd need upwards of 100 runs.
If you want to try it all in one run, it'll take more than that.
Well if nothing else, a strike against multiple passes is that it's going to be single segment. Even segmented, I still don't see the point though, because Providence takes awhile to get and it'd be more time than it's worth to make multiple passes.
(My current goal is to get 3 weapons and a Silver Eye and Gold Eye and then providence out in less than 10 minutes. Actual time is probably going to be 4-5 minutes
That does raise the question as to whether the AC is actually going to save me any time.
Ah. We'll see, I guess.
To be honest, I doubt it would if it takes 10 minutes. 4-5, probably. Since my estimate is more like 3 minutes (unless I'm forgetting something like you wouldn't need to go to Gruberik otherwise, and then maybe you need to watch a boat ride or something), that's why I think the AC is worth it.
Although, I don't know how you'd feel if after all this, it turned out that the AC was actually a waste of time. =/
Our opinion: The people involved with that site seem to be full of themselves. Also, many of those are widely known.
All I really wanted to say (if it hasn't been said already, sorry) is that 1P should actually be faster than 3 players because you can run and use up your charge, switch, use up the charge by running again, etc. Although I would still understand the appeal of a 3 player TAS, especially since with all the poles and bouncy guys and things that sync all the players to one location, you can have 2 players goof off or kill enemies and not lose speed.
Gorash: That's counterintuitive but interesting. Since the text advances by character, do you have an explanation? (like the box only lets you advance after a certain delay, which would sync the name differences) Obviously for text that advances by pages ("Purim sees the reaper" in battle, or other games altogether), the name length wouldn't matter.
I like this series a lot so I have full support behind this. Although it is not very much like Ogre Battle and more like Fire Emblem or Shining Force. Also, saying it is "much better" than Ogre Battle is almost like pure flamebait in my mind. =p Actually I am surprised Phil didn't get mad at you.
Oh yeah and I think that "certain someone" is banned already.
Hrmmm. If he got it in less than 7 hours, then he's a much more patient man than I am. I can't count the number of times I've run into a wall because the timing on the controls has such a small frame of opportunity to stop running. Not to mention running into enemies.
My current estimate for finishing this run is 5 hours, give or take 10 minutes. The revised run is probably going to be more like 4.5 hours.
I dunno, I got the feeling that guy was blazing through it as a WIP. He did the whole thing in just a couple days I think, and to be quite honest, the decisions he made that he posted were very baffling (picking up the speed ring, doing the world's most difficult puzzle, things like that). Well, with your TAS estimate, I suppose I will revise my console estimate to 5 hours then. ;)
quietkane wrote:
Yeah, definite jealousy, there. It is beyond the realm of human possibility to truly optimize the AC. I'm just hoping that the final run will have a good enough run of it. (My current goal is to get 3 weapons and a Silver Eye and Gold Eye and then providence out in less than 10 minutes. Actual time is probably going to be 4-5 minutes, and the chests I pick up are unlikely to include both jewels, but I can probably get 3 weapons and one of the Eyes.)
That sounds like a good thing to shoot for.
quietkane wrote:
A console run probably won't be that bad, since you aren't going to have the pressure to make it perfect. I'd say it wouldn't take more than 6 or 7 tries overall to get a good random placement for the cave, and you probably won't need more than one equipment-collecting run before you can take the whole thing. The aggravating part about a console run is that, like I said, it's hard to efficiently avoid enemies, and the Dragons on the lower levels are going to seriously scare you.
No, trust me, it's going to be a total pain, and it'll take a lot more than 6 or 7 tries (probably 100-200 at least), although most of the tries are going to be on the shorter side since after getting suboptimal luck early, it's time to reset. Oh, and since it's from start of gift mode in the title screen (i.e. without carrying anything in), I would do this in one pass. Any more would waste time of course. Maybe you were thinking it would be easier if I would be carrying equipment in.
quietkane wrote:
Sorry I don't have a WIP for you. I'm having to re-take one my finals on Tuesday, and I really need to study for it. (So why am I posting here instead of studying? Well, I needed to get on the internet to look something up anyway, and I figured I'd take a study break while I was here.)
Wow, a retake would be completely demoralizing. I have a long story from last week that I can sum up as: I misremembered my final exam time (!) and ended up going over to the professor's house to take it later that day. Of course I would have been extremely upset if I weren't allowed to take it at all, but even if I were allowed to take it say the following week, I would have still been pissed off since it would still be just hanging there.
Say, has anyone tried avoiding fights on the outer world with save-reset-load? Seems to me that it would be faster than running. Does the SMV format allow recording reset or power switch?
Ha, I was just thinking about the implications of power cycling for FF4-6. In 6 it takes far too long after reseting to be able to load your game to save any time I think. In 5, it would be worth it, except that the place you'd use it the most would be the overworld, and just bringing up the menu resets the step counter in FF5's overworld. It still might be useful for certain save points, but I figure only if they're directly in the path and if there's a boss coming up, you can't just walk to the boss without incident. Whoever's running should test, since it might be a close call. In FF4, it would help a lot on the overworld. I might be doing this one myself, and if so, I was fully planning on using it.
There was some talk about this game at SDA before, and I have personally beaten this game with savestate abuse at level 32, so I know a TAS would be able to be low level.
Besides Kevin, Hawk is also very obvious. The last character should probably be Riesz. Actually, I don't think it's very tough to decide. Also, it's not as bad as you suggest with 7! and such since many of those combinations are obvious throwaways and you can just follow the natural progression of the days.
About class change, I don't really see the point in a TAS. Definitely the 2nd one is not needed. Maybe the 1st one is to speed things up since your stat ups get caps.
i don't really know, but i'd think an rpg would be fairly straightforward since you have a clearly defined goal for each event. i did think about playing with rudra no hihou but when i realized i have to learn large amounts of japanese gibberish i put that one aside (i think the game could make an entertaining run though, and i might still consider it).
Well many games, like say SMB, are level independent. RPGs tend to be harder because it's more like one big level, meaning the planning is harder since you have to make some decisions way in advance. About Rudra no Hihou, I've played it a little, but I couldn't keep interested at all. I do know you can basically cast all the spells from the beginning, but other than that I don't know how runnable it'd be or even if the game is any good.
spineshark wrote:
at any rate, learning how to use the tools isn't too hard/time consuming on its own, but using them effectively is much harder, as far as i can tell.
Well um, using them effectively is implied in using them at all! ;)
spineshark wrote:
haa, i hate advance. so boring. and the battle theme sounds horrible (i programmed the melody into my cell phone for a ringtone! haha).
You think it sounds bad in all versions or in the GBA port? I've never been the biggest Uematsu fan, but FF4-6 is definitely his best work IMO, and back in the SNES days he was rather good compared to most composers. His overworld themes tend to be very good. Yeah, just like Mitsuda, battle themes are pretty much a big weakness, but I thought the ones in this game were decent.
spineshark wrote:
i feel the same way, and all i could reccommend is, don't rush it. i am thoroughly amazed at the people who can work what seems like very quickly to me.
Heh, the last thing I am worried about is rushing it. =p
spineshark wrote:
i would certainly enjoy a tas on this game (i look forward to watching your run as well). working on a tas and then viewing your handiwork is...fun. but also quite strange. it's odd to go through things over and over again, very slowly, then see them happening so deftly and perfectly. i don't know if you'll like it, but if it means there's a run on this game here, then i'd be happy. =)
I don't know if anybody else feels the same way, but watching my own console runs already feels quite strange. ;)
Right, and besides, why don't we move away from this "star" overanalysis to some "moon" analysis? Better than blue stars, red stars, flashing stars, glowing stars!
How can you say no to that smiling face?? That just screams "Hello, I am a noteworthy publish. Please watch me. 3-up!"
Wow, I was pleasantly surprised at the quick reply. This is great news! And you have good answers. ;) I realized I was confusing myself and some of the comments I made were meant for a console run, whoops.
First off, when I got 12 minutes, that was without save states, so it was supposed to be like a console concept demo. I didn't fight any unnecessary fights except for the frog right in front of the Big Catfish, to get to level 5. I got the Insect Crush because conventional wisdom is that you would. I didn't get the Catfish Jewel, although even in a console run, you would want to redo that segment until you got it. I used Boomerangs on basically everyone: 2 to kill the Lizard Man, 4 to kill Big Catfish, and 1 for each fight with multiple regular guys.
For a TAS, I'm don't think you'd want the Insect Crush either, but I'm not 100% sure. Also, I agree you would skip Stronger, unlike what a console run would have to do I imagine. Even if you engage minimal battles, Boomerangs are useful to kill all the multiple regular enemy fights quickly, and you'll need them at lower levels vs the early bosses.
Ah, of course Tia attacking was manipulation. The switching positions is better, yes.
quietkane wrote:
Dragondarch is going to have his work cut out for him trying to run this game. The controls are very imprecise, and the fact that he can't efficiently avoid battles is going to hurt a lot. I have sincere doubts that this game is doable within 10 hours without tool-assistance. Good luck to him, though.
(and I've forgotten whether SDA goes by in-game clock or not. If they do, then he needs to spend time in the Ancient Cave. Time spent in the AC doesn't count for the main game's clock.)
Actually, getting under the 7 hour SDA soft limit won't pose any difficulty, but Maxim's fast movement and trying to optimize the paths/puzzles will.
The guy who I said wasn't doing a very good job (27 mins compared to my 12 mins to get Tia and save), completed his test run with 6:37 at the post game timer. I forgot what you estimated earlier for a TAS, but my (I think it's conservative enough) guess for a console run of Lufia 2 would be between 4 and 5 hours. I've been working with DTK on his 4 Lufia 1 test runs and he has that down to 5:21. I think the real version will be close to 5 hours, and I'm pretty sure that Lufia 1 would take longer than 2, so there goes. ;)
SDA uses real time from gaining control after starting a new game, to final hit against the last boss. AC is a definite "no" for a console run. Ah, this might make you jealous. =p
Oh yes, maybe I haven't told you, but after Dragondarch does his Lufia 2 run, I am going to do a Gift mode console run of the AC from scratch. =) Actually, that smile might disappear after I try it and pull all my hair out from the randomness.
I've watched all the star runs and completed every game on the recommended page except for Ninja Five-0 and Sonic Advance 2, where I messed around a bit.
That being said, flagitious's Umihara Kawase is one of my favorite runs on the entire site, and I would be strongly against unstarring it. I agree with Gradius and River City Random though. Back a page, I also recommended unstarring (I'm ready for the crazy stares) Bionic Commando and LTTP. If need be, of course... it's not 1-to-1 add/remove but eventually some will need to be removed.
Now, on the non-issue of having two different colored stars, which I think could be confusing, I created a SMW moon icon. 18x18 PNG with alpha transparency, like the others, and I used PNGOUT to compress it all the way down.
What do you think?
Catastrophe: That was a great explanation, thanks. Sounds good and I learned some things. ;) Erim does seem to pick up fewer chests and win fewer battles, I think. I could be wrong.
Erim: I made the same mistake as Catastrophe with the Sprint Shoes and the unequipping.
if i were doing it this way, the answer seems obvious to me. you kill off the lemming as early as possible, so he doesn't drop anymore. that'd save a second or so on each level.
Well actually, there might be better situations. Let's say on a level you have to go "the long way" to save all the lemmings, but if you use an Exploder in a certain spot, then you sequence break if you will, and have a much shorter path. Something like that.
spineshark wrote:
i rather like the idea of maximum (for this purpose, "100") percent, as opposed to a "best rank" run. especially if some of the tribes don't give you much/any leeway anyway, i don't see what's the point of killing off 10 extra lemmings (or however many it would be) to save 45 seconds.
Well, whatever's most impressive is best, and I don't know what that is. But all else equal, I do agree with you. ;)
I haven't confirmed this, but it is possible that you would need 38 to 40 lemmings at the end of each world to get the gold medal, as in most of the worlds you can save all 40 lemmings, but in some you will have to give up a couple, but typically the surviving lemmings at the end would be from 38 to 40. So that may be about it to get the gold medal.
Well, I remembered that the requirements for gold are different for some tribes than others. Then, I said what the hell and just tested. With Shadow, if you let even one lemming die, you get silver.
Sketchie wrote:
Anyway, if you plan to kill off the remaining lemmings to make the level go faster after getting an apporitate number of lemmings in the goal, just use the ever-so-handy Exploder skill. :P Or fall off a cliff, jump into acid, get into a trap, whatever, to save time. You could even use the flying gadgets (Magic Carpet, Jet Pack, Super Lem, Icarus Wings, and so on) and force the lemming into a pit, trap or acid to kill them off faster.
Well, there is no shortage of ways to kill guys if you want to (armageddon being the obvious and always available, and just escaping is best in versions that allow that). The trick would be to find a level where sacrificing a lemming would allow for a faster solution. That's what I meant.
Bag of Magic Food wrote:
Wait, so now we're making gold medals the goal rather than a nice side effect of 100% possible saving?
Actually, we could do either, and it's up to spineshark. It would be a lot easier to plan with 100% possible so the levels would stay independent. Both would be impressive in different ways, and there's really not going to be too much of a difference either way, so it's a judgment call.
I guess it could depend on how tricky you have to be to do 100% possible. I don't really remember. Although, with more trickyness, that leads to more gold leeway, which would also lead to more room for sacrificing timeshaving. Well the first step regardless is to figure out exactly how many you need in each tribe.
Well I don't think any of us feels like any of the current star movies are short of amazing, or that it even really matters, but clarification can't hurt.
You might like full runs, but some people like me actually like the shorter runs better since there's less repetition, usually more route planning (not in this case), usually more difficulty (not in this case either though), and so forth. Also it would take less time for a newbie to watch (who would be sampling a lot of the star page perhaps), and might be less confusing. "Why didn't you take the Star Road shortcut?" etc.
This is probably just my personal bias against 100% runs though. On SDA, maybe part of this is sucking up, but everyone spooges over marsh's warpless SMB3 and barely acknowledged Sleepz's warp run (other than alleging that it was a TAS I guess). I actually liked Sleepz's run more because it was MUCH better at levels they both did, and with that, there was even the issue of endurance with warps vs no warps. Here, endurance isn't an issue, so I can't think of any argument other than "I like to see all the levels." Which is valid of course.
Phil wrote:
Oh and I won't post or read this thread anymore after today. Those are just my final words.
Is Phil really quitting the scene? He seems so involved and dedicated, I find it hard to believe. And even if he's not just threatening, people quit making console speedruns at least and come back all the time. Anyway, if it's true, that's rather sad as his ability is absolutely top-notch, but maybe he is fed up with something because it seems like he's getting into fights in every other thread he posts in.
Anyway, he's kinda cast a long shadow over this thread, because everyone knows that if he's doing it, they should just let him do it as he'd probably do a better job. =p
Actually, I'm probably more concerned about the length, which leaves more room for things to go wrong. Is it really that tedious? Can you manipulate when you have your next random fight with say, when you end the previous fight? Because if you can't, the walking around would be pretty straightforward I think. And in battle, would you just pause and then unpause and see if that's a critical, and if not, frame advance once and try again? Maybe I'm naive about the difficulty. I certainly have been for console runs (ICO "one weekend" estimate -> actually over 3 months). =/
I think probably the biggest hurdle to me would be learning all the nuances of TAS making. Of course, this would probably apply (to different extents) no matter which game I did.
Oh and I would definitely not use the GBA version since I don't like it, don't like remakes in general, and I don't know it nearly as well as FF2 US. This is assuming of course that I do decide to do this. I guess I'm just scared at how long this would take, since I have more console runs I want to do, games I need to play and catch up on, and also my thesis to work on, heh. I mean, I think DW2 was done in 2 days or something (!) but that must have been a busy 2 days, and I think it would take me a lot longer. I think in several weeks, I will start this, see how it goes, and then decide whether or not I want to continue.
I hope quietkane is doing ok, but it's time for me to give this a bump of love!
Anyway, so while I was going back and watching all the TAS WIPs I hadn't before, I watched this one of course, since I probably care about this the most. Well, I'll write down what I thought of.
- more boomerang... Actually on SDA, some guy was doing a Lufia 2 test run (and not doing a very good job at all). He got Tia and then went back to town and saved in 27 minutes. I thought this could be much improved, so I tried myself, legitimately on emu. I got 12 minutes, with the Insect Crush (since you'd want it if you were to keep on going), and basically I boomerang'd the crap outta everyone. With the WIP, I took control and walked to the save cleric, and got the very start of 17 minutes. This makes me rather sure that the beginning could be done better. I fought less fights, used more boomerangs, and didn't get Strong. However, I don't think it's even worth buying Strong, just waiting until you can get Stronger. Potions and Boomerangs are quite good enough I think, and faster.
- avoid more enemies (i.e. by walking) instead of shooting? I don't know if this actually could have been done more, but given quietkane's optimization on the movement, puzzles, and everything else, maybe not.
- less wasted action in battle (i.e. with Tia)... A lot of people forget about this. It's generally faster to defend with bad offensive characters than it is to attack with them for relatively small amounts of damage. Maybe Tia's attacks are needed? I don't remember.
- Tanbel Southeast Tower twice? The first time is to get some overpowered equipment. I'm generally a big fan of RPG run minimalism, but possibly this was ok since you don't have regular encounters in Lufia 2 and you run like the wind. Still, a whole traversal of a tower is a lot. I shudder when I see non-obvious chests being picked up, sorry. =p
- nice use of Drowsy vs Idura! I didn't know you could do that. That's the hardest battle in the entire game too, so I don't know if maybe it could be done elsewhere but it isn't worth it, or what. I went back and tried it myself, to see if it was a weakness, but you have to luck manipulate to get it of course. =p Well, it's no Dragon Warrior I suppose.
Again, I hope you're doing well and everything is peachy keen as the ofay kids say. Even if you have lost interest, which would be sad but understandable, you should know that when Dragondarch does Lufia 2 for SDA (it's next in the queue), your work thus far will be very helpful to him. =) (And me too, since I'll be helping him.)
I wrote the ultima garden thread at SDA, so I knew all about those things, including the PS2 fast loading option. I also read that you were just doing this for fun, with the translated version and all that.
I guess what I am saying, is that I am not making a comment on your speed vs ultima garden speed. (That is fine because you are doing us a great service by providing proof of concept.) I am making the comment that 2 minutes does not seem to me like something you would say "wow" about like genji did. He only has 10 posts so I figured maybe he doesn't understand TAS's enough. For example, marsh at SDA was confounded that hero's CT TAS was 30 mins faster, when actually, only being 30 mins means that his run was very good. Also, Strangeness in the Mischief Makers thread here was impressed that the first WIP someone did for the first level was better than the console time at all.
I'm thinking it would probably be easier to do a non RPG for a first TAS, but if I decided to do FF2, I think I'd be able to do a very good job without too much hassle since I already know the route I'd take. Also it's apparently easy to manipulate luck in battle with pausing.
If you are referring to NES FF's, I'm not interested. For NES games in general, I'm actually thinking of doing Dig Dug 2 and might start a topic. But I have no idea if people would find that interesting or not, and my goal is to get published. This is a bit frightening with non-marquee games, since great runs IMHO like the El Viento (which was even obsoleting a previous version by the same author), took half a year or so to get published. At least with FF2, that would get published like a madman.
Well I see that despite all the support you are getting from people in this thread, who would love to see this, I guess you've decided not to do this. =( Definitely it wouldn't be the most boring movie on the site to most people, but perhaps you are afraid that it wouldn't be published since it's not famous enough? Breath of Fire isn't exactly obscure but whatever.
Anyway, I finally went and watched the whole thing. I was actually a bit disappointed that it really seemed like you knew what you were doing, because I guess in the back of my mind I was wishing there would be more room for improvement. I did take some notes though, and I'll share them, even though some of them are probably things you'd clean up for a real version anyway, and some of them are probably flat out wrong.
- wipe quicker when you want to death abuse, instead of from lots of health
- why fight more than one trip of K Sludges? Maybe you didn't want to get stuck later, but it seemed to me like the leveling was generally overkill. For example, you mauled Deathevan. And large benefits of higher levels (being able to run better, less encounters) matters less in a TAS.
- why not restore out of battle during the 2nd K Sludge trip?
- low AP instead? I said this earlier, that it might even be beneficial to let Ryu die at times to keep his AP low. Although the more I watched, the more it seemed like you did this correctly, but part of that is because of the relatively higher than I expected levels, you needed dragon moves less.
- more just defending instead of doing time wasting moves? This is actually a pretty big one maybe, because sometimes you do things like have Jean attack for 8. Ironically, this is the next big strategy timesaver for the console BoF3 run as well.
- no steps before rolling? I noticed a couple times you walked a bit before rolling with Rand. I wonder if there's something I don't remember about rolling that would necessitate that. Maybe whether or not you fight is based off the tile you start rolling, or something like that.
Anyway, after all that, I have an idea that I am dearly hoping you will consider. While maybe you think a BoF2 TAS would be boring, it is exactly the type of thing that we would like to see at SDA. I think you posted in that thread (I have been AWOL; I will post this there as well), but perhaps you didn't think to do this because it was already in Dragondarch's queue. Well somebody would have to ask him, but since his queue is longer than the one at the DMV, I am rather sure that he would cede that one to you and let you take care of it instead if you wanted. You don't know how much some of us would want to see a console run of BoF2, so if you are up to it, this would be great indeed. =)
Well now that Tilus is doing FF5 and hero wants to do this, I guess I will throw out that I would like to try this over the summer if nobody else has gotten around to it. I have decided that I will do my first TAS then, but I'm still trying to figure out which game. I'm very familiar with this one, but then again that might not be the best idea since all the strategies are already engrained in my mind.
SaxxonPike, are you referring to normal play, watching a speedrun, making a TAS, or what? Making the TAS for this has to be a lot more boring than the others, and playing is debatable, but IMHO watching a TAS/speedrun of FF3 is much more boring than FF2 or CT. It's not even close. Logically this is the case because it is more broken and degenerate, e.g. Joker Doom.
I mean, realizing of course that they are all very boring compared to watching games that aren't RPGs. ;)
http://tasvideos.org/ updated.
