• Plays on hardest difficulty
  • Manipulates luck
  • Hits things with a fist
  • Takes no damage
  • Very slight glitch abuse
  • Neutron Bombs everyone (except Heihachi)

History of this run

I got a funny idea one day that concerned fighting game speed TASes, mainly that... why the ********** aren't we playing using the quickest possible settings?! Aside switching difficulty to hardest, we've always just used the default settings of the game, like keeping 2 rounds per match. So, this paradigm changing TAS (cough) instead chooses to only play 1 round per match, cutting the painful repetition by half. Seriously, if any other game in another genre asked you "How many times do you wanna beat this guy up?" would you choose anything else than the quickest answer?
Aside winning a lot of time by fighting less, this TAS also uses improved tactics compared to McBobX's original speed TAS of Tekken Advance.
I did a test speed TAS of Tekken Advance previously, trying to gather some opinions before deciding what to do next. This current TAS improves it by a little less than 3 seconds. If you're curious, the test TAS can be seen here.
Also, the re-record count is a bit misleading since I made this run using the test TAS movie file. About 2000 of those re-records come from the test TAS, but hey, it's been a process of learning... so in a way those re-records are just as official as the rest.

Basics, and... why Paul?

So, your usual Tekken Advance round can be broken down to two interwoven objectives that you must complete as quick as possible.
The first objective is simply doing enough damage to reduce the opponent's HP to 0. In Tekken Advance your opponents have varying amounts of health. Paul, Xioiuya, Nina, Law and Heihachi have only 140 HP and are the squishiest. Gun Jack has 168 HP and is the toughest single character to take down. The rest have 151 or 152 HP. So this forces you to switch up tactics a bit depending on your foe.
Second objective is manipulating the enemy to willingly open themselves up for a beating. This is a tricky part since the A.I. in this game directly reacts to your behavior, and can't be manipulated pre-emptively before a round starts. This is to say, you cannot manipulate your foe without losing time, and you only have a finite amount of opportunities to manipulate the foe before you've lost too much time! (Too much in this context means that your TAS begins to stink.)
I chose Paul as the fighter because he has a very good damage over time ratio, with his Burning Fist being the game's fastest and (almost) most powerful super attack. Outside that one extremely fatal attack, he is also a pretty versatile character with a whole catalogue of decent attacks to start the round with. Versatility is good since then you have more leeway navigating around the A.I. when it is feeling particularly reluctant to go along with your schemes.

So, getting started...

Every round you start with some starter attack or move. This move varies depending on how much HP the enemy has and how the A.I. is feeling that particular moment, and its purpose is to either bring the enemy down to around 120 HP so that they can be neatly finished with Burning Fist OR to manipulate the enemy to use their own super powered attack during which they get full counter hit damage, killing them at best in one shot... and speaking of countering...

Countering

If you attack an enemy during their attack, it is a counter hit and they receive extra damage. On paper you get a 150% increase in damage... but in effect you only get a 120% increase, with a good amount of damage variance thrown in as well. So basically, Paul's Burning Fist that initially does 100 damage and SHOULD do 150 damage in a counter hit, in most cases only does 120 damage or whereabouts.
It seems that the counter hit damage varies depending on what attack the enemy is doing, and also how far into the attack animation the enemy is. The enemies' own super powered attacks (equivalents of Burning Fist) are the riskiest and therefore give the best counter hit damage as well. Like mentioned, in theory you could one-shot almost every foe in this game if you got them to do their own super attacks like buffoons at the start of the round... this phonemenon can be seen in the playaround TAS against Nina... but in practice this is highly rare. So rare I couldn't manipulate it to happen in this run at all, except in a very imperfect form in the Gun Jack fight. Maybe some other person who understands A.I. at a program level or can do some bot manipulation could make it work. And this takes me to the subject of...

A.I. Manipulation

A.I. manipulation plays a very important role in the TAS. From the start of the round, you can do the following things to manipulate A.I.:
- wait
- quickstep forwards or backwards
- interrupted quickstep
- sidestep (takes about 6 frames)
- fake-jump (hold U+D for two frames, then let go) (takes about 6-9 frames)
- Sway forwards or backwards (takes about 10 frames) (Paul-only)
- jump forwards or backwards (takes a lot of time)
- jump and do air attacks (takes a lot of time)
- do any other attacks (takes a lot of time)
Crouching doesn't seem to help much in manipulating the enemy. Sidestepping to either direction is also counted as the same thing as far as the A.I. is concerned, so that limits things a bit.
So, trying different outcomes to manipulate the enemy A.I. is a tangled mess, especially once you start combining techniques like interrupted quicksteps and fake-jumps. I tried to be systematic about it, but overall my most important goal was simply to improve the test TAS times with new tactics instead of finding out the cryptic RNG combination that allows to one-shot everyone. So once I got a solution to a round that was an improvement and something I was satisfied with, I kept it and didn't look back. Anyway, getting back to those starter attacks...

List of starter attacks

Paul has the following worthwhile options for starting the round with.
- Basic kick. 30 DMG, counter hit is 40-45 DMG. Knocks back enemy. Fairly slow since the attack lasts 50 frames. One perk to this starter is that since it doesn't require any d-pad input, you are able to do it 4 frames earlier from the start of the round than any attack that does require d-pad input. Also, the damage is just perfect for most enemies. So it has a situational use.
- Shredder, i.e. two jumping kicks. Damage seems to vary a lot, averaging at about 28 DMG if both hits connect. More if it is a counter hit. It's a fairly slow starter, but sometimes against weaker foes you can only do the first kick and it will be enough (like against Xioauio in this run). Has situational use. One good thing about this is that the attack contacts with your foe fairly quick, around 15 frames after starting input.
- Double Hop Kick High. The same thing as Shredder, although with a different input, a different (useless in this case) follow-up after the two kicks and slightly more damage. Didn't use it in the run, actually.
- Bone Breaker. Two hits that deal a total 34 DMG (or more if counter hit). Otherwise a pretty cool starter that knocks the enemy down on his back, but it's one of the slowest as well. The whole move sequence lasts about 75 frames. These damaging starters sometimes allow a tactic against weaker foes where you don't have to manipulate a Burning Fist counter hit to kill them since the starter does enough damage already. This was seen in the test TAS but not here.
- Gut Buster. Two hits that deal a total of 40-44 DMG (or more if counter hit). Same deal as with Bone Breaker. Fairly damaging, knocks enemy on his back, fairly slow.
- Glitched jump A. 25 DMG, counter hit is 37 DMG. This is an attack all the characters in the game have. It's a relatively quick attack at only 33 frames, although setting the glitch takes 9 frames. A counter hit will knock the enemy on his back. Basically you have to hold U+R for 9 frames, then do the jumping kick. The character does the jumping kick without ever leaving ground, which makes the normally air-only attack more wieldy. This is one of my favorite starter attacks because of the speed, glitchiness and surprising deadliness. However, the kick has a bit of a cooldown as well. Some frames after the move the game thinks your character is "landing" and you are unable to do other specials since the game will prefer to do "rising" attacks instead. But you can walk during this time, so it's not all completely wasted time.
- Quick PK Combo. Or some of the other similar combos Paul has. These strings that start with a jab contact at 11 frames, allowing you to sneak in a punch between enemy combos. Overall following the jab with the kick can make this combo fairly slow, but the fast start and enough damage ensure that it has a situational use at least.

Other things

Many of these starter hits knock the enemy back, so you may have to catch up to him again by running a bit. The good part in this is that running allows many possibilities for influencing the A.I. behavior... the bad part is that it's of course slower this way. Sometimes the A.I. is friendly enough to actually roll towards you after being knocked back, becoming easy pickings for Paul's Burning Fist when they rise up.
Needless to say, 140 HP characters are usually the easiest to face since you can take them down with less hefty (and usually faster) attacks. Some unique strategies are possible with these guys, for example you don't always have to manipulate two counter hits to knock them out.

Wait, were you going for in-game time again?

Well, long story short, I was planning to go for in-game time in this TAS, which is to say: fight all the fights to the bitter end. But after witnessing the chaos of the TASVideos IRC I became shocked, and in a fit of temporary insanity played for fastest input instead (sorry Spikestuff). The good news is that I managed to do early input end without touching the round timer at the options menu, saving some extra frames in the process!
Heck, now I'm just being polemic for the sake of it. Hey, if people don't like this ending to the run, it can be changed so that the run goes for in-game time instead. So not a big loss either way. It's just... hearing more opinions about it would be appreciated.

Round by round comparisons:

Round 1, vs Jin. Test TAS time 2.60. New TAS time 2.31. (0.29 secs ahead)
Glitched jump A counter hit. Manipulate Burning Fist counter hit.
Round 2, vs Nina. Test TAS time 5.23. New TAS time 4.40. (0.83 secs ahead)
Glitched jump A. Manipulate Burning Fist counter hit.
Round 3, vs Yoshimitsu. Test TAS time 7.58. New TAS time 6.80. (0.78 secs ahead)
Sidestep, basic A. Manipulate Burning Fist counter hit.
(I felt like I had very bad luck in this fight since I actually lost a little time.)
Round 4, vs Law. Test TAS time 9.95. New TAS time 8.81. (1.14 secs ahead)
Glitched jump A counter hit. Manipulate Burning Fist counter hit.
Round 5, vs Gun Jack. Test TAS time 12.50. New TAS time 11.20. (1.30 secs ahead)
Hammer of the Gods. Manipulate power attack from enemy to quick kill with Burning Fist.
Round 6, vs Xivaaouyuy Test TAS time 14.88. New TAS time 13.15. (1.73 secs ahead)
Shredder first kick counter hit. Manipulate Burning Fist counter hit.
Round 7, vs Hwoarang. Test TAS time 17.35. New TAS time 15.31. (2.04 secs ahead)
Basic A counter hit. Manipulate Burning Fist counter hit.
Round 8, vs King. Test TAS time 19.85. New TAS time 17.53. (2.32 secs ahead)
Quick PK Combo counter hit. Manipulate Burning Fist counter hit.
Round 9, vs Heihachi. Hard to compare since my test TAS didn't go for in-game time...
But the early input sequence is a few frames faster at least.

Closing notes

Now, the only question that remains is... how did Paul's right arm become so strong? (wait, don't answer that)
Thanks to Spikestuff for making a Lua script for this game... which I admittedly didn't use in the making of the final TAS, but it's the thought that counts. Also, sorry to Spikestuff for betraying his hopes more than once.
Thanks to everyone who believes in fighting game speed TASes although they are nauseating to watch and have little intrinsic worth.
Thanks to someone else too.
This TAS is surely not perfect. So, feel free to try and beat it (but not by too much... that would be humiliating).
Okay, that's it.

feos: Judging...
feos: Input ending "too" early is a heated discussion subject, but we have already solved the issue. The author chooses to leave it end early, I accept it to Vault.
As for the tag battle mode, if it's available from the start without any passwords or SRAMs (unless the password is surely known beforehand), it can be a separate category, even in Vault. But I can't guarantee it can obsolete the any% run, will depend on reception.


Site Admin, Skilled player (1254)
Joined: 4/17/2010
Posts: 11477
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
It's advised, if the author wants to change the ending, that he does so before the run is judged. But he's free to use whatever ending he likes himself. http://tasvideos.org/JudgeGuidelines.html#InputLength
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Joined: 7/10/2007
Posts: 280
The hypocrisy shown by the members on this site with respect to fighting game TASes is ridiculous. Voted yes, this is exactly what I want to see with a fighting game TAS. I don't care that it's repetitive.
ALAKTORN
He/Him
Former player
Joined: 10/19/2009
Posts: 2527
Location: Italy
dunnius wrote:
MUGG wrote:
I don't like this kind of "input end" where the character stands around for another minute before the game ends. I prefer fastest real time
I agree. It was this reason that I voted no on E.T. I didn't like the standing around at the end. It detracts from the entertainment.
What entertainment? This is a fighting game speedrun, it’s not supposed to be entertaining. Check out Cooljay’s run for entertainment. I found it entertaining exactly for the clever trick, but I’m just trying to say that it’s weird to hear about entertainment in a fighting game speedrun aimed for the vault.
Spikestuff
They/Them
Editor, Publisher, Expert player (2643)
Joined: 10/12/2011
Posts: 6439
Location: The land down under.
TASVideoAgent wrote:
feos: Judging...
Well enjoy seeing the words "accepting to vault" without writing anything else for this game and judging between ingame for fighters vs TAS time for TASing. According to what feos has written within this forum he was a-okay with the 1 minute of killing the point of a speed fighter TAS. THIS IS SO AWESOME TO SEE (sarcasm)
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account. Something better for yourself and also others.
NhatNM
He/Him
Experienced player (718)
Joined: 6/17/2009
Posts: 600
Location: Vietnam
*clap* *clap* (applause) :D
A man come from Vietnam My YouTube: https://www.youtube.com/c/NhatNM/playlists
Editor, Experienced player (885)
Joined: 1/23/2008
Posts: 529
Location: Finland
I think many people here have strong preconceptions of what a fighting game TAS should be like... To start with, for the longest time we had complaints about fighting game speed TASes in general - as if fastest theoretical completion time was not a worthy consideration in this particular genre. But... why wouldn't it be? Fast forward a year or so. Now that fighting game speed TASes are slowly becoming more accepted, we get to bickering about the settings. It's been been customary to do fighting game speed TASes at default settings (other than difficulty), at the expense of pure speed. Again, why? ...It's like fighting games exist in a small bubble separate from the other game genres and their established TAS rules. Now, I can see some defensible arguments on the both sides of these questions (especially against early input end). But a large part of the discourse seems fueled by bias. Perhaps feeling implicitly betrayed since the default, age-old format of a fighting game has been tampered with? It's clear that this kind of 1-round TAS looks and feels different compared to previous fighting game speed TASes... in both good and bad. But in the end it doesn't have to look sensible or entertaining, it's going to the Vault anyway, most likely. And pure speed should be the thing that matters. In other words, let's discard the bias and embrace the cold, heartless objectivity of speed instead. For great justice! --- Regarding early input end, here is a Strawpoll that decides what happens. Vote yes or no to early input end, and hopefully we'll get a clear consensus on what people actually prefer. And yes, this is a real democratic choice - whichever option wins (in, say, two days time) will be in the final run. Also, don't be lame and try to unfairly skew the poll results to one way or other. Otherwise it is a step away from true democracy towards a-holecracy. And you will feel horrible guilt about it for the rest of your life (I hope). http://strawpoll.me/2822641
Joined: 12/10/2006
Posts: 118
One could easily have two versions. One for each option.
Spikestuff
They/Them
Editor, Publisher, Expert player (2643)
Joined: 10/12/2011
Posts: 6439
Location: The land down under.
thommy3 wrote:
One could easily have two versions. One for each option.
Only TAS which I can think around those lines.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account. Something better for yourself and also others.
Site Admin, Skilled player (1254)
Joined: 4/17/2010
Posts: 11477
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
Spikestuff wrote:
According to what feos has written within this forum he was a-okay with the 1 minute of killing the point of a speed fighter TAS.
No, it's you a-okay with how a run loses 1 minute of the game finishing by itself. Also, why do you love absurd so much? You have a thread about that matter, where opinions exactly halved, and you are defending your personal preference with indicting me for bias, ignorance or something else?
Spikestuff wrote:
Well enjoy seeing the words "accepting to vault" without writing anything else for this game and judging between ingame for fighters vs TAS time for TASing.
Why so butthurt? The "ingame for fighters" issue is completely new to me. Was there a thread about how it's preferred? Maybe some guideline I'm stupidly missing? I just want to remind, that leaving the ending type to the author's choice is a guideline, since both options are equally preferred.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Cooljay
He/Him
Active player (397)
Joined: 5/1/2012
Posts: 468
Location: Canada
haha that ending. You gotta use a Low Heihachi. Other than that it was a meh.
ALAKTORN
He/Him
Former player
Joined: 10/19/2009
Posts: 2527
Location: Italy
Please don’t change the ending. Tekken has a unique blocking system, and abusing it this way is wonderful. Changing it to be the same as all other fighters is plain stupidity.
adelikat
He/Him
Emulator Coder, Site Developer, Site Owner, Expert player (3573)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
I think the end input choice is correct in this situation. Normally this kind of tactic isn't well received and not preferred. It adds far more actual time, of uninteresting gameplay. But in this game it works, because it adds some much needed variety. At last there is a fight that is not punch, fireball, win. (It goes without saying though, I voted no, as I was not entertained)
It's hard to look this good. My TAS projects
Joined: 3/9/2009
Posts: 530
AKheon wrote:
Again, why? ...It's like fighting games exist in a small bubble separate from the other game genres and their established TAS rules.
The main answer to most of your questions is because TASing fighting games for speed looks far FAR less impressive than regular play (assuming players of at least moderate skill), and absolutely less so than things like combo exhibition videos which people have been doing at regular speed for ages. Furthermore, there are generally only two skills on display in a speed oriented fighting game TAS, damage planning, and AI manipulation. Neither are complicated and neither will change much from round to round. Damage planning is even more trivial for fighters than most games since there are entire databases dedicated to frame exact info and timing already out there, and infinitely so when there exist broken gimmick moves like the super punch here, Yoshi's seppuku, Soul Calibur's ring outs, etc. tl;dr Speed oriented fighting game TASes look worse than regular play and are accomplished mostly through endurance, not ingenuity.
NhatNM
He/Him
Experienced player (718)
Joined: 6/17/2009
Posts: 600
Location: Vietnam
Tangent wrote:
Soul Calibur's ring outs
Eh, you make me remembered Soul Calibur can work this last input trick. then should I do new version for Soul Calibur with this trick? (just make a hit on enemy then be down for dodge all attacks) :D
A man come from Vietnam My YouTube: https://www.youtube.com/c/NhatNM/playlists
Editor, Experienced player (885)
Joined: 1/23/2008
Posts: 529
Location: Finland
The early input end really does divide opinions. At this moment, 45% want no early input while 55% are in favor of it, with only two votes difference... (22 total votes)
xxNKxx wrote:
should I do new version for Soul Calibur with this trick? (just make a hit on enemy then be down for dodge all attacks) :D
I think the rules state that just changing ending input and submitting it as a new run isn't justified. The run should have other improvements as well...
Tangent wrote:
Speed oriented fighting game TASes look worse than regular play and are accomplished mostly through endurance, not ingenuity.
Hmm, perhaps. To me it depends on the game, though. Maybe we need more fighting game speed TASes that rely on precise glitches to get through the game really fast... *remembers Mortal Kombat 4 (GBC)* ...or maybe not.
MESHUGGAH
Other
Skilled player (1918)
Joined: 11/14/2009
Posts: 1353
Location: 𝔐𝔞𝔤𝑦𝔞𝔯
I agree with Spikestuff about versus (fighting) games should be use in-game time since everyone use it for (RTA) WR's. "If you attack an enemy during their attack, it is a counter hit and they receive extra damage." So, why didn't you used Yoshimitsu? It would deal ~90% + 20%, an instant kill on each round. Is there any reason besides "every tekken uses yoshimits"? edit: obvious yes vote casted
PhD in TASing 🎓 speedrun enthusiast ❤🚷🔥 white hat hacker ▓ black box tester ░ censorships and rules...
Spikestuff
They/Them
Editor, Publisher, Expert player (2643)
Joined: 10/12/2011
Posts: 6439
Location: The land down under.
MESHUGGAH wrote:
So, why didn't you used Yoshimitsu? It would deal ~90% + 20%, an instant kill on each round. Is there any reason besides "every tekken uses yoshimits"?
Yoshimitsu doesn't have the move Spinning Harakiri in the GBA version. In any case he's actually a slow and least recommended character for the GBA version.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account. Something better for yourself and also others.
Editor, Experienced player (885)
Joined: 1/23/2008
Posts: 529
Location: Finland
Yoshimitsu is not that great in Tekken Advance. His power attack (stab) only does 90 DMG, with 135 DMG on a perfect counter hit. It's both weaker and slower than Paul's Burning Fist. Combined with the fact that Yoshimitsu lacks as many fast round start options as Paul, manipulating enemies with him is more of a pain and fast kills are less likely. And like Spikestuff said, Yoshimitsu's superpowerful gimmick attack (hara-kiri) does not exist in this game.
MESHUGGAH
Other
Skilled player (1918)
Joined: 11/14/2009
Posts: 1353
Location: 𝔐𝔞𝔤𝑦𝔞𝔯
Spikestuff wrote:
Yoshimitsu doesn't have the move Spinning Harakiri in the GBA version.
I'm referring to this one https://www.youtube.com/watch?v=HaAyO4E8Kbs#t=65 Maybe the damage only increased because of tag battle mode? Never played this version.
PhD in TASing 🎓 speedrun enthusiast ❤🚷🔥 white hat hacker ▓ black box tester ░ censorships and rules...
Editor, Experienced player (885)
Joined: 1/23/2008
Posts: 529
Location: Finland
Yeah, that's the "stab" power attack. Two of the three stabs on the video are counter hits, and at best case a single stab can put an enemy into critical status. For now I maintain though that Burning Fist is better... Characters do in fact have less health on Tag Mode... hmm. I hadn't even realized this since it was so long since I last used that mode. Good call. I wonder if this changes things... EDIT: I sense a new TAS may be incoming. At least if Tag Mode is possible to use without it being a different category altogether.
Joined: 3/9/2009
Posts: 530
AKheon wrote:
Tangent wrote:
Speed oriented fighting game TASes look worse than regular play and are accomplished mostly through endurance, not ingenuity.
Hmm, perhaps. To me it depends on the game, though. Maybe we need more fighting game speed TASes that rely on precise glitches to get through the game really fast... *remembers Mortal Kombat 4 (GBC)* ...or maybe not.
It's not only fighting games. Sports games are a good example as well. You could certainly finish most faster or slower by adopting certain strategies which are impossible in regular play, but that's absolutely nothing like what is considered high level play. For example, in a football game, you could always rush so the clock never stops, then manipulate defensive penalties for automatic first downs so the AI never gets the ball to stop the clock either. Then score at the last second. You'd finish the game much faster than regular play, but that's nothing even close to what high level play looks like. Or in a soccer game, pass it back and forth so there are no stoppages for goals.
MESHUGGAH
Other
Skilled player (1918)
Joined: 11/14/2009
Posts: 1353
Location: 𝔐𝔞𝔤𝑦𝔞𝔯
I understand your point Tangent, but... "certain strategies which are impossible in regular play, but that's absolutely nothing like what is considered high level play." Why an "impossible strategy in regular play" isn't considered as a high level play?
PhD in TASing 🎓 speedrun enthusiast ❤🚷🔥 white hat hacker ▓ black box tester ░ censorships and rules...
Joined: 3/9/2009
Posts: 530
MESHUGGAH wrote:
I understand your point Tangent, but... "certain strategies which are impossible in regular play, but that's absolutely nothing like what is considered high level play." Why an "impossible strategy in regular play" isn't considered as a high level play?
Mainly it's because high level play against AIs doesn't exist. It's common knowledge that the AIs are full of exploitable behavior even in regular play, so relying on cheap/cheesy moves that take advantage of how poor they are, especially doing one move over and over again, is 'scrub' play that anybody is capable of, not expert. Or like in the football example, the primary goal of the game isn't speed, so strategies dedicated to finishing quickly are nonsensical.
Editor, Experienced player (885)
Joined: 1/23/2008
Posts: 529
Location: Finland
So I took Tag Mode for a whirl and started a TAS of it. By the end of the third fight I'm about a second ahead of this run. So, in case Tag Mode and Arcade are analogous enough to both be useable in any% TAS - something I'd like to hear feos' opinion of - maybe this submission could be on hold for a moment until I get this new version with its improvements finished. (Or should it be a new submission altogether?) One thing that makes Tag Mode pretty different (other than different tactics) is that there is lag that you need to worry about... guess having the extra HUD sprites on the screen drags the performance down a bit. So you can't mess around after the fight because even that wastes extra time. With enemy HP lowered, the action is so fast right now that many of the normal options you'd consider to manipulate the foe or start a fight with are simply too slow. So TASing becomes in a way "simpler" since you have less options to consider. However, at the same time lag adds a whole new layer of problems to things. Fastest possible in-game time solution may not be the fastest input file time solution any longer. All this is fascinating and horrible at the same time...
ALAKTORN
He/Him
Former player
Joined: 10/19/2009
Posts: 2527
Location: Italy
I’m not sure how “Tag Mode” could count as an any% completion. Surely a TAS should use the main game mode, Arcade Mode.