1 2 3 4 5 6 7 8
Joined: 7/7/2007
Posts: 161
I'm skeptical of the Calculator approach and grinding in general. I hypothesize that manipulating dodges with mantles, perhaps assisted by Faith augmentation, is sufficient and will be a better time value. Of course both methods should be tested, but I'm a bit surprised at how the unanimous consensus seems to be that pushing toward Calculator is the fastest approach. Maybe in an unassisted game like Chessjerk's, but infinite critical hits coupled with infinite dodging is too strong to ignore. Plus, man is obtaining Calculator tedious. I can't imagine making the audience sit through battle after battle on Mandalia. Automated manipulation is the core of my approach. I can't imagine brute forcing this game. I need a new computer, though. Even with scripting it still takes an afternoon to run a bunch of trials, and I need to use my computer for other things. I'm trying to hold out for Sandy Bridge; hopefully "Q1 2011" means January 1st.
Joined: 8/17/2007
Posts: 41
I don't know how much free time i have but what i do have is a beast of a computer. I'd be happy to run trials if you provide me with the Lua's and criteria when i have time. I wanna see this made so i can make time to help you.
Wren
He/Him
Joined: 4/11/2005
Posts: 196
Location: Michigan
Yeah, I also have my doubts about calculators saving any time at all. After all, there is this little thing I like to call a counter attack. I can only imagine how cool it would be to see one guy run into a group of 4 only to counter every single blow sent his way. It'd be like watching Neo whoop some bro-bot ass... only in super slow-mo. Not to mention this would issue a nice amount of extra JP. This run is a nightmare to plan but the eventual run will be spectacular to anyone who played the game. Good luck to whoever runs this.
<i>A little rudeness and disrespect can elevate a meaningless interaction to a battle of wills and add drama to an otherwise dull day.</i>
Former player
Joined: 3/27/2010
Posts: 270
Well maybe I could shed some light on this.. As I've been playing the crap out of FFT waiting for the next release of the rr emu xD ... Ramza's sign = Cancer (Possible to abuse luck to make the generic we train to be Cancer as well?) You don't get JP from counters, only when you have your own turn (CT is at 100), so just no on that... We could simply just obliterate the first battle in as little time as possible, all with criticals... After the first battle is over and we actually have control of Ramza in the main first real battle... The chemist in this battle has (as far as I know, I've actually spent 4 hours or more just in this battle, just hitting him and letting him heal and repeat) unlimited potions... So we need to simply raise Ramza's JP up til he has enough JP for yell... In a non TAS run, you would actually train most of the generic characters.. But since this is a TAS run, we really only need to train 2 (the ones with the best compatible zodiac sign) ... So as far as I could care, we let one of the chemist characters survive, the others... We kill, and let them crystallize, that way, it speeds up the battle (less turns), and we don't have to worry about kicking them out of the party later (every second counts), plus.. Ramza would be using wish on the chemist we let live, so the crystals he gets, will heal him. Ramza, should already knows dash, and I believe theres a chance that he'll know throw stone, if not, there might be a chance to get it from the enemy's crystals... Frankly, as for money, I've found, you really don't need to waste it, or spend it.. Just use it when you need it. Anyways, when the chemist has unlocked Wizard, and Ramza has enough JP for yell. We can end that battle, in as little moves, but the best number of moves for our benefit. That's just an overview of it, if you want to go into more details, then I could help plan each move by move. Also, remember to kill Delita xD... We don't care about him anyway, until later ;) After that, we take the accessories (like battle boots) (if any :p) , off of the other generics, and then kick them out of the party.. There is never going to be a reason to save, so that, also saves time. So, we should have 3 people in our party at this point, Ramza, Delita, and the chemist. Buy yell for Ramza, leave Delita alone, use JP scroll glitch on the chemist job of the chemist.. Buy only these skills, as these will only be necessary in the run... Potion (done have), Pheonix Down, throw item, , Elixir, ether. auto potion. Next, go to the priest job, and JP scroll glitch it, and buy everything that is necessary, which is all cure spells, all raise spells, esuna, holy, regenerator (for later)... Put in your abilities, for the chemist .. 1) Black Magic 2) Item 3) Auto Potion 4) Throw item 5) -blank- For Delita leave everything as is, since he is a guest, being that we can't control him, we need him to cause as much damage as possible before dying, without dying something else.. However, buy things like throw stone for him, if he doesn't already have it. Next is Ramza, change him to a chemist, buy the necessities and yell if not already (listed above), have his abilities like this 1) Item 2) Guts 3) Auto Potion 4) Throw item 5) -blank- Now we need to buy a generic that has cancer, if it's impossible to have one start with you... I personally think to save us hours of headaches constantly watching a cutscene over and over and over til we finally get it right, this would be much easier. We just need to make sure that it knows throw stone ;) ... Then the rest is simple, we continiously train it, until mediator is unlocked.. With talk skill learn invite as well as the skills below... We use invite on enemies that we actually can't kill in 1 turn (if possible). Now that you have the basic idea of what you need, you're going to have to watch Chessjerk's entire run ... Every part, that way, you can get your ideas out of your head as to what you think is best (unless it really is xD), and look at it from a speed running point of view. We have not a single second to waste, so instead of sitting back and leveling, we level stuff and progress through the story simultaneously ... Back to the JP scroll glitch anyways, The main problem with using the glitch is that is does NOT raise job levels, so we need to have Ramza Yell his way up to level 4 Priest, level 4 Wizard, level 3 Oracle, and level 3 Time Mage to open up Calculator, and here comes the tricky part. Once we finally do get calculator up, Ramza needs to have lare, Quick, Demi, Demi 2, and Teleport, and teach the generics Bolt 2, Bolt 3, Frog, and Magic Attack UP, as well as Preach and Mimic Daravon for the Talk Skill generic. We'll also need half of mp bought so it's easier for people to train/kill everyone. (get from Summon skill I believe). In all due time of course, but as for now, we have 1 level 1 character with 00 exp, who we need to train to be the mediator, one character who is going to be a time mage, Delita, after the next battle .. Algus, and our wizard. I believe we should talk advantage of this battle, NOT like in Chessjerk's run, since we have inhuman possibilities :P with TAS♥, but just enough that we can unlock some classes, and beat the battle in the best way possible to our benefits. Afterall, this game is about tactics, not about speeding it and being dumb... We'll take every single battle in the game to our advantage, absolutely NO random battles, and buy the best items that cause the most damage, with the best evasion rates :) .. Eventually, we'll just have Ramza power abuse calculator everything... I believe it'll take less than 2 hours to unlock calculator this way (or should atleast)... Then by time we unlock it, we should be at the 2nd Lucavi battle (Wiegraf isn't it?) :p .. We'll that's my input on it. ^^
Joined: 1/4/2011
Posts: 35
Glad to see this as a potential for a TAS. I wanted to point out that I did do a run for SDA without using Math Skill. I'm not sure if it has been referenced yet, but here is a link: http://www.archive.org/details/FinalFantasyTactics_Ninjas_655 Some people that I've talked to definitely found this run much more interesting (albeit much slower than my Math Skill run, obviously).
Joined: 7/7/2007
Posts: 161
Thanks, chessjerk. I liked the earlier run and wasn't aware you made a new one. I will check it out.
Joined: 5/11/2006
Posts: 71
Just dropping by with an opinion or two.. Counter and (its been a while, so forgive me if Im wrong) Hamedo would obliterate everything. Screw JP when you just crit-counter punched someone's face. Counter works on monsters, Hamedo works on humans (hits before attack and cancels theirs? I think? been a while). Mechanically, theres so much to abuse in this game. Martial arts + anything that increases PA + counter/hamedo can result in OHKOs. Speed can be abused to no end, and the JP scroll glitch to unlock some fun stuff too (teleport! 100% success rate in a TAS :P). I'd think that calc's would just suck overall. Too long to unlock, too slow (literally, ingame speed), too long to train up. On assassination battles, they'd be even worse. Personal note: I'd love to see someone use throw stone to knock someone off a ledge and kill them that way. Sorry for the disjointed thoughts, I haven't been in this thread in a long time and am feeling renewed interest in this. Someone mentioned a new version of a recording emulator coming? Whats that about?
Samus taught us that a girl doesn't need brains to be successful. Brains are giant, evil, and vulnerable to missiles.
Joined: 5/11/2006
Posts: 71
Forget everything I said. After watching the current SDA run, theres just no real way to compete with a calculator run. Sure, you could save time at the start by just pressing forward (instead of JP farming), but you'd lose the advantage quick. Turn 1 > Ramza w/ MP switch > Flare CT5 > everything is dead in one shot. No amount of critical hits, teleporting, counter, etc, can compete. Sadly, its not so much a matter of "calculator is substitute for TAS precision", but more like "calculator skills nuke the entire field in 1 shot". And frankly, I'd never want to see a Calculator TAS. It'd just be too boring, wouldn't have any entertainment value, and would just be X minutes faster than the current SDA but mirror it. Having said that, I'd love to see a "no calc" run, or a "single class only" or "single CHARACTER (ramza) only" run. But those would probably be shot down as arbitrary, since they aren't as fast as a calc run. The only other option, which I'm sure wouldn't be allowed, would be a TAS run of the 1.3 patch version of this game. For those who don't know, it completely changes the game and makes it near perfect balanced. It removes the JP scroll glitch, removes "broken" skills, makes enemies very well balanced (enemies with actual reaction abilities? gasp), and a whole bunch of other stuff. Basically, you couldn't cheese your way through it if you tried. But, it's a patch, and probably not well enough known by people to understand why its a GOOD patch, not just some random hack. Too bad, because IMO a patch run would be amazing since it actually makes the game hard (or at least, forces you to THINK instead of just cheezing your way through).
Samus taught us that a girl doesn't need brains to be successful. Brains are giant, evil, and vulnerable to missiles.
Joined: 8/23/2008
Posts: 417
Callmewoof wrote:
The only other option, which I'm sure wouldn't be allowed, would be a TAS run of the 1.3 patch version of this game. For those who don't know, it completely changes the game and makes it near perfect balanced. It removes the JP scroll glitch, removes "broken" skills, makes enemies very well balanced (enemies with actual reaction abilities? gasp), and a whole bunch of other stuff. Basically, you couldn't cheese your way through it if you tried. But, it's a patch, and probably not well enough known by people to understand why its a GOOD patch, not just some random hack. Too bad, because IMO a patch run would be amazing since it actually makes the game hard (or at least, forces you to THINK instead of just cheezing your way through).
Personally, I'd say this is the way to go--it wouldn't be a TAS of the original game and would probably warrant a publication under the "hacks" section of the site, but fuck it. Having played 1.3, I can promise anyone who takes it on as a TAS that it's going to have significantly higher entertainment value and be far more challenging/impressive than calculator facerolling. Someone should do this.
I will not use self-reference in my signature.
Experienced player (961)
Joined: 12/3/2008
Posts: 940
Location: Castle Keep
The big problem (I have?) with this is simple, if we are to allow that then its open floodgate of hell for having _any_ game patched and rerun. Imagine mario nes without pipe glitch and stuffs like that, i think it would become dull run and not very impressive/fun. Of course you guys are not speaking of mario, you have good reasons ect... Just saying this should not set a precedent, having this as a hack category sound reasonable solution imo. (this doesnt mean we forcably need first a no-hack run, i think) Note: AFAIK, the game dont go over the "intro" (first battle)? even in psxjin (im not totaly sure on that last bit) ? kinda settle the issue if the game is not supported at all
Joined: 7/2/2007
Posts: 3960
I'd say it more opens the gates for difficulty hacks. Imagine a SotN hack where everything didn't die in 1 hit and you couldn't spam powerful attacks every which way, for example. Even if the hack incidentally also fixed a bunch of glitches it'd still be acceptable (assuming this hypothetical hack were well-made), though it certainly wouldn't obsolete the unpatched runs. I don't think a hypothetical Mario hack that fixed pipe glitches would be acceptable if that were all it did; the result would just be a Mario run without the use of entertaining glitches.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Emulator Coder, Skilled player (1311)
Joined: 12/21/2004
Posts: 2687
For a while I've thought that hacks/patches would be a neat way around the issue of wanting to make something like a "glitchless" run but not being able to define what that means without it being just an arbitrary set of restrictions. With a fixed version of the game, you wouldn't have to "hold back" while TASing and it could simply be a regular TAS of that version without needing weird goals and rules. As for calculator vs. non-calculator run of this (unpatched) game, I did some really rough timings and calculations and it looks like a non-calculator run would need to maintain an average time of at most somewhere around 90 seconds per battle in order to have any chance of beating a calculator run. (That's timed from when any player or enemy character's status stuff appears at the bottom of the screen to when the status display last goes away at the end of the battle.) That sounds really hard, but considering that almost half of the required battles can be beaten by killing a single enemy, and that criticals and misses can be manipulated, and that the JP scroll glitch can still be used, I wonder if that isn't as crazy as it sounds.
Joined: 8/23/2008
Posts: 417
arukAdo wrote:
The big problem (I have?) with this is simple, if we are to allow that then its open floodgate of hell for having _any_ game patched and rerun.
That's a reason I've read before, but we don't actually get a floodgate of submissions now anyway, and I don't think there's a long line of people sitting around waiting to TAS hacked versions of games. Besides all that, if it provides more entertaining runs, would it be so terrible to have to deal with a selection of bad submissions alongside it? Just like SotN, the challenge of this game is severely undercut by its own mechanics, which hampers the technical impressiveness of a speed run. Unlike SotN, though, there's no "let me win that in less than 20 minutes" run to make impressive use of glitches in a reasonable timeframe. Instead, we watch a calculator murder everything in one turn for a few hours, which just isn't quite as fun.
I will not use self-reference in my signature.
Joined: 12/27/2007
Posts: 50
Location: UK
Acheron86 wrote:
Just like SotN, the challenge of this game is severely undercut by its own mechanics, which hampers the technical impressiveness of a speed run. Unlike SotN, though, there's no "let me win that in less than 20 minutes" run to make impressive use of glitches in a reasonable timeframe. Instead, we watch a calculator murder everything in one turn for a few hours, which just isn't quite as fun.
Goddamn, that was well put. A non-calc run would indeed be much more entertaining, but as to deciding on a original/patched version, I'm undecided. In a way, there's something not quite 'wholesome' about them. But if it really is good a patch as you guys are saying... who knows.
Joined: 7/25/2007
Posts: 109
Acheron86 wrote:
Personally, I'd say this is the way to go--it wouldn't be a TAS of the original game and would probably warrant a publication under the "hacks" section of the site, but fuck it. Having played 1.3, I can promise anyone who takes it on as a TAS that it's going to have significantly higher entertainment value and be far more challenging/impressive than calculator facerolling. Someone should do this.
Not a fan of 1.3 myself. They pretty much just nerfed anything that was remotely useful to hell and back, and gave the computer opponents enough buffs to make the Civilization series jealous. In fact, I saw someone on the team state that the original intent was mere rebalancing, not balancing the stick in the other direction. There's another hack out there called LFT which I always wanted to try (which changes things up to make them more interesting/useful/better), but chronic procrastination. :/ Then again, seeing 1.3 absolutely decimated would be amusing in itself, and LFT is still being tweaked to this day, and given the "latest version" rule/guideline it wouldn't work.
Joined: 7/30/2011
Posts: 129
Location: Watching a TAS in the basement...
I would certainly love to see a TAS of this game. I can't do it because the game doesn't run right. Gafgarion's turn comes in the first battle, but he doesn't do anything. If somebody could help me get this to run, I may be able to TAS it. Otherwise, I do have a lot of knowledge of this game. I could help with route planning.
I am the future ruler of the world! My forum: http://elderyoshisisland.forumotion.com/
Joined: 8/15/2010
Posts: 3
TASManiac wrote:
I would certainly love to see a TAS of this game. I can't do it because the game doesn't run right. Gafgarion's turn comes in the first battle, but he doesn't do anything. If somebody could help me get this to run, I may be able to TAS it. Otherwise, I do have a lot of knowledge of this game. I could help with route planning.
I actually came here today having the same problem! Tried different emulators and different rom downloads. After hours I feel like I've lost some IQ! I want to playaround with some non-calculator routes that will require ridiculous luck manipulation while remaining creative....if I could only TAS my way past the first turn of the first battle LOL
Former player
Joined: 3/27/2010
Posts: 270
Ephew Anjrmahm wrote:
TASManiac wrote:
I would certainly love to see a TAS of this game. I can't do it because the game doesn't run right. Gafgarion's turn comes in the first battle, but he doesn't do anything. If somebody could help me get this to run, I may be able to TAS it. Otherwise, I do have a lot of knowledge of this game. I could help with route planning.
I actually came here today having the same problem! Tried different emulators and different rom downloads. After hours I feel like I've lost some IQ! I want to playaround with some non-calculator routes that will require ridiculous luck manipulation while remaining creative....if I could only TAS my way past the first turn of the first battle LOL
The emulator can't detect the event triggers that make the guest moves, so this game is untasable at the moment (afaik, that is the reason why). However, once the update comes to let this game by tasable, it'll definitely be on everyone's list. It shouldn't be a hard fix to do, as emus such as epsxe, can run the game just fine. Unfortunately though, the run will be a calculator run... As much of a disappointment that is to some, it is the fastest way. But it doesn't mean non-calculator runs can't be made. It just wont be displayed on the TAS list.
Joined: 1/4/2011
Posts: 35
I for one would definitely like to see a TAS for both calculator and non-calculator runs. On another note, I just finished improving my calculator unassisted run for SDA. The final time was around 4:22 (about a 22 minute improvement). I guess a TAS could get this under 4 hours, maybe?
Former player
Joined: 3/27/2010
Posts: 270
chessjerk wrote:
I for one would definitely like to see a TAS for both calculator and non-calculator runs. On another note, I just finished improving my calculator unassisted run for SDA. The final time was around 4:22 (about a 22 minute improvement). I guess a TAS could get this under 4 hours, maybe?
I'll have to watch that soon. The TAS would definitely get under 4 hours (perfect crits and what not), but (without watching/guessing), I'm assuming anything under 3 hours is far fetched. So I'm guessing somewhere from 3:30-3:50 would be the time for non calc. Just by better optimization you saved 22 minutes, and the TAS would more than likely save another 22 minutes, but it require an insane amount of discussion between people. Hopefully they can fix this issue as it'll be a TAS of the year type thing :p
Joined: 1/4/2011
Posts: 35
Has anyone actually come up with a battle by battle strategy for a TAS? Or is it too soon for that, since apparently you can't get past the opening battle? My thoughts: For a calculator run similar to my unassisted run, the only place where any significant time can be saved is in Chapter 1 (better starting JP in various jobs, faster JP grinding, etc.). Assuming a similar setup to mine, once you get to Math Skill, every battle is over in one turn (two for Zodiac beasts, four for the final boss), and since I already do that, the only significant time saving after Chapter 1 will be in not taking the time to save, which costs me about 5-10 seconds per save. For a non-calculator run, would it be a similar run to my 6:55 Ninja run (I know this run is very unoptimized, but I'm not sure by how much)? Or, would the idea be to get to Monk (which sadly can't be scroll glitched), and learn Counter or Hamedo as fast as possible. The problem would lie in whether Hamedo-critical hits can kill everything at a low level and without good equips that raise physical attack.
Former player
Joined: 3/27/2010
Posts: 270
chessjerk wrote:
Has anyone actually come up with a battle by battle strategy for a TAS? Or is it too soon for that, since apparently you can't get past the opening battle?
Well I've got bits and pieces of strategies, but nothing majorly like step by step. As soon as the devs fix this glitch, I'm fully willing to co author this with someone. But until then can't do much. But I have, with save states, messed around and find how crits/miss/etc can be manipulated easily enough, for the non-calc run. I can't really say much on the other stuff you said until I can get around to watching your full runs again, which will take a few days on my part, but someone else might can answer.
Joined: 1/4/2011
Posts: 35
Do you mind sharing what those bits and pieces of strategies are? I am curious. If not, that's OK.
Former player
Joined: 3/27/2010
Posts: 270
chessjerk wrote:
Do you mind sharing what those bits and pieces of strategies are? I am curious. If not, that's OK.
Yes ofcourse. Just really the obvious things though. -Crits always, unless unnecessary. -Attack enemies based on CT. (Make it so the enemies don't really get turns). - Once we get agrias, abuse crush punch (atleast I beleive that's the one), so all hits are instant death. - Manipulate enemies to walk into death traps. Such as abusing flare/death in the non calc run. - Using our guests as an advantage and killing them when necessary to save turns/time. That's just the basics though really. I have a good strategy for the sand rat level which consist of abusing who goes first, blocking them all inside, then using either flare/death. Though (even though it wasnt flare/death, was pretty much the same), i tested this on v1.3 so it works out pretty good. think the spell was cyclone that I used, that ended up killing them all very fast. If we can, we can also abuse charm in this.
Joined: 1/4/2011
Posts: 35
Torn338 wrote:
Yes ofcourse. Just really the obvious things though. -Crits always, unless unnecessary.
Definitely.
-Attack enemies based on CT. (Make it so the enemies don't really get turns).
Makes sense.
- Once we get agrias, abuse crush punch (atleast I beleive that's the one), so all hits are instant death.
The main problem here is Agrias is slow, so usually she will go after the enemies, whereas a Ninja would normally go before the enemy. Unless you can somehow get her an Excalibur or the perfume that grants Haste, I don't see this as a viable strategy.
- Manipulate enemies to walk into death traps. Such as abusing flare/death in the non calc run.
I'm assuming you mean having the enemies walk into your spells (and not using the traps on certain battlefields that inflict Death Sentence)? Does that mean you are thinking of using a party with magic users? I'm not sure how that jives with your previous point about not letting the enemy get turns, since it's very difficult to go first and cast a spell before the enemy moves.
- Using our guests as an advantage and killing them when necessary to save turns/time.
Of course.
That's just the basics though really. I have a good strategy for the sand rat level which consist of abusing who goes first, blocking them all inside, then using either flare/death. Though (even though it wasnt flare/death, was pretty much the same), i tested this on v1.3 so it works out pretty good. think the spell was cyclone that I used, that ended up killing them all very fast.
Summon Magic should be even better in that battle. Manipulate high Faith/good Zodiac compatibility on all the enemies and they should all die in one summon.
If we can, we can also abuse charm in this.
Abuse charm? Are you talking about purely for entertainment purposes, since why charm an enemy when you can kill them instead? General battle strategies are good to have, but I think we need to work on some specifics. Some of them have been tossed around in this thread, but I don't think any conclusions have been reached. I think we need to start with: 1. What is the goal? Lowest time without Math Skill? Lowest time by any means? Lowest time with the most entertainment? 2. What party will be used to accomplish that goal? Solo Ramza would likely be fastest most of the time (mostly due to only having to set up one character for battle), but it might not always be feasible. EDIT: Something else that just occurred to me: How long are all of the unskippable cutscenes in this game? That might give us an idea of the potential time for this run.
1 2 3 4 5 6 7 8