Idle Flinch - More Proof



  • Hey all, I almost never play MaA so I loaded up frams and gave it a go playing MaA in FFA. I managed to get a 1:1 “flinch working properly” and "idle flinch bug " one after the other. There is an explanation in the video, but I’ve included it below as well.

    Is there any word about a fix in the upcoming patch?

    _hCc98EIVGY
    http://youtu.be/_hCc98EIVGY

    A video showing a bug that occurs in Chivalry: Medieval Warfare.

    Explanation of the bug:
    All attacks are assumed to be slashes.
    The Dane Axe has a 0.45s windup animation and a 0.625s combo.
    The Grand Mace has a 0.525s windup and a 0.75s combo.

    This means there is .625s from when my first attack releases to when my second attack passes it’s “windup”.

    After attacks pass the “windup” they can not be stopped even if the character is hit.

    If a character is hit (and not in post windup) he suffers a 0.2s “flinch”.

    So, an example if I am swinging a Grand Mace and I hit my foe (and he is not in the post wind up) and combo into another slash. The timeline would be as follows:

    (A): Knight initiates the attack
    (A+0.525s): Knight’s attack is now post windup and uninterruptable.
    (A+B): Knight’s attack lands on MaA. He initiates his combo. MaA is now “flinched”.
    (A+B+0.2s) MaA recovers from flinch.
    (A+B+0.75s) Knight’s combo is now post windup and uninterruptable.

    We see that there is a 0.55s gap between MaA recovering from flinch and the knight passing his combo. Plenty of time to dodge, throw up a parry, even throw a stab (0.4s windup)

    –-----

    You’ll see from 00:15 - 00:17 I attack the Vanguard twice with the Dane Axe. The first attack (0.45s windup) is followed by a combo (slash + slash). And we see that the Vanguard gets flinched (notice his head) after both hits.

    At 00:18 I start the first attack on the Knight and he is post windup and we see his flinch (head cocked forward). The bug is shown at 00:20 where he is clearly hit before his windup is complete, but somehow able to continue his attack after being hit.

    Following the timeline, he is hit and goes into flinch (0.2s) as I start my combo (0.625s) leaving him with 0.425s to act between my hits. His windup is 0.525s and should be unable to go off between my attacks. (His proper play would be a kick, parry, or repositioning)



  • From what I saw, against the knight at the elast.

    You flinched him to start, and hit him before he started his attack.

    Then he hit you. I dont think the animation for it was synced though, as he went through it too quickly.

    That makes me think hes just lagging.

    Kick laggers.



  • All too familiar with that. That second hit should have surely flinched the knight. Definitely needs a fix.

    Seems like people will blame lag for every “bug” but also everyone seems to have a vastly different idea on what an acceptable latency is.

    I don’t think this is a lag issue as it occurs even with low ping players.



  • Yeah finally an example by video, I have that happen to me a lot.



  • There is one problem with your explanation. When you said that there is a 0.55s gap between flinch and knight’s post-windup, there most likely won’t be enough time for a stab with Dane axe. Damage doesn’t occur instantly after the wind up finishes, instead there is a 0.5s release time (for Dane axe). This means that your attack most likely won’t be in time to interrupt the windup, unless you were at hugging distance, attacked as soon as possible, and the combined latency (of both you and your enemy) is under 150ms (0.55-0.4 = 0.15). The animations don’t represent everything when latency is involved, and when you are dealing with openings in timing as small as this something as small as 0.1s delay caused by latency may make all the difference between a flinch or a trade.

    Looking at your video, it appears to me that when you hit the knight for the second time he hasn’t started his windup at all. Flinch only occurs when a person is attacked during windup, not while he is idle. It’s not so much of a bug as it is a design flaw. I don’t know for sure, and mods can probably answer you better regarding when it may get changed. I know that the devs are aware of this, but it may or may not be part of the changes in the upcoming January patch.

    As for how he managed to hit you between your combos, the answer is simple. He didn’t start his windup and wasn’t flinched! Your combo takes 0.625s to windup, while his windup is only 0.525s. If both attacks started at the same time, it’s not surprising that he would hit you first considering how close you were (ergo takes a lot less release time to be hit). Higher latency will also cause his attack to land slightly earlier than the animation suggests. While the difference may be as minor as 0.1 or 0.2 seconds, as said earlier in such small openings those differences matter.



  • I still don’t get it, watching that video twice all I see is a bad MAA player blaming the boogie man for getting killed.

    PS: what I saw was working as intended.



  • @GhoXen:

    Flinch only occurs when a person is attacked during windup, not while he is idle

    I’m almost sure this isn’t the case and can be seen by the vanguard hits just a few seconds earlier.

    –-

    @Blackjack:

    I still don’t get it, watching that video twice all I see is a bad MAA player blaming the boogie man for getting killed.

    PS: what I saw was working as intended.

    Not sure if mad or jelly.



  • @BOT:

    @GhoXen:

    Flinch only occurs when a person is attacked during windup, not while he is idle

    I’m almost sure this isn’t the case and can be seen by the vanguard hits just a few seconds earlier.

    Hm, they must have changed things since I started? I did take a break from the game for awhile. I just tested it out, and you are completely correct. I withdraw all my comments regarding the validity of the idle flinch bug.

    The calculation of timings in your explanation are still not that correct, but that is irrelevant to the issue seeing the video. The only way the knight couldn’t have flinched is if the attack was already in the release state, which simply wasn’t the case.

    Just for reference though, was the knight in question high in latency? High latency is always a potential cause. If you want the proof to be flawless, it would be a good idea to eliminate high latency from the possibilities by recording only “idle flinch bugs” occurring under near-optimal latency (80ms or less).



  • @BOT:

    Not sure if mad or jelly.

    You flinch him, then hit him again just before his flinch ends, like 0.2 sec later he starts attack, you die because you are not good and it looks like you failclicked the third combo checking the vid more than twice now.

    Working as intended, by the way, you think by now that this bug could be reproduced with 2 people on a server with no latency issues and repeating that 10times for youtube already standing still side by side :?

    Not mad just saying it looked ok to me and something that happens often with 2 blows too fast or screwing up your combo, like spamming dagger for instance, flinch from what I have seen is a set limit. Funny you write “idle flinch” in the game when you get axed.

    PS: If that is not how it is suppose to work then I would just go with archer and the dagger and kill every thing, every time. flinch, flinch, flinch, flinch =dead



  • @Blackjack:

    You flinch him, then hit him again just before his flinch ends, like 0.2 sec later he starts attack, you die because you are not good and it looks like you failclicked the third combo checking the vid more than twice now.

    That cannot happen. It’s impossible for him to land a second hit on the knight within 0.2 seconds, and it didn’t happen in the video. The knight did not flinch, nor wait 0.2 seconds after he was hit the second time, since he started his attack immediately after taking the hit, when there should have been a 0.2 sec idle flinch. If he did indeed flinch on the second hit, and started his windup 0.2 seconds later, the MAA’s combo attack would have landed first, yet the MAA was flinched during his windup.

    Not mad just saying it looked ok to me and something that happens often with 2 blows too fast or screwing up your combo, like spamming dagger for instance, flinch from what I have seen is a set limit. Funny you write “idle flinch” in the game when you get axed.

    There is no such thing as a “set limit” for flinch. Using a dagger you can always flinch a target (tested on friendly bot), as long as they don’t parry.

    PS: If that is now how it is suppose to work then I would just go with archer and the dagger and kill every thing, every time. flinch, flinch, flinch, flinch =dead

    No, a flinch is only 0.2 seconds. There is no attack in game that is that fast. Even dagger stab combo takes around 0.75 seconds to deal damage. There is still 0.55 seconds of free time, enough for a parry, but not enough for a kick (against a fast weapon like dagger at least).



  • @Blackjack:

    PS: If that is not how it is suppose to work then I would just go with archer and the dagger and kill every thing, every time. flinch, flinch, flinch, flinch =dead

    Unless the enemy parried, which would be the appropriate thing to do after being hit and seeing a combo coming.

    Kicking is also an option against some slower weapons (0.2 flinch + 0.35 windup)



  • That is a pretty clear case of flinch failing for none of the “usual” reasons like hitting at the end of a recovery or during release. As has been said already, the only explanation excepting a bug would be latency - but that can’t be the case here either. If you watch closely, the first hit flinches late in the Knight’s windup, while the second hit lands right at the start. The only way latency would factor in would be if the Knight had such a high ping that his windup animation was playing when he was in release client-side - clearly not the case if a late swing flinched but an early swing did not. This is further confirmed by the knight’s attack hitting about when you would expect it to in the animation.

    Also as mentioned before, hitting an enemy who’s already in the flinched state is not possible for one player as no weapon has a combo time of .2 seconds. Hitting a player who is idle also inflicts a flinch. So there is no logical reason why the attack did not flinch in this case. Call him a bad player for slash spamming all you want, this could’ve happened to anyone - of course, it doesn’t happen to very good players specifically because they combo feint every single attack, as it’s the only way to avoid trading hits with the wonky flinch.



  • @SlyGoat:

    Call him a bad player for slash spamming all you want, this could’ve happened to anyone - of course, it doesn’t happen to very good players specifically because they combo feint every single attack, as it’s the only way to avoid trading hits with the wonky flinch.

    So basically, a “good player” is someone who’s discovered a way to effectively circumvent a poor (“wonky”) game mechanic?

    Sounds more like exploiting weaknesses in the game than skilled playing to me.

    I don’t know about you, but I’d personally rather have a game requiring proper skills, than a game based on finding ways around inaccurate game-mechanics that result in the controls behaving in a unpredictable manner. In fact, a game claiming to be skill-based shouldn’t allow for these sort of things, or it would clearly not be skill-based in the true sense of the word.

    Also, those who claim this is “not a problem” or that it’s “working as intended” are obviously the very same players who’ve have had the poor taste of exploiting this “wonky” mechanic to their own advantage and would most likely not be anywhere near as successful should they be forced to rely on actual proper skills instead.

    EDIT: Hacks are of course a completely different matter, but as long as these issues are in the actual game and is not the result of modifying the game by hacking, then they certainly need to be addressed by the devs. Also, lag only goes so far as an explanation. It seems pretty clear that neither lag nor hacks can be used as an excuse for all of these incidents.

    -E-



  • Is everyone going to ignore how fast the first swing of the knight is compared to the second?

    Its really clear hes lagging and he just got desynced.



  • All I know is that what happens in that video happens all the time, I accepted that as mechanics and working as intended and I adjusted my game to it long ago. I called him bad because in that video I saw it coming, I would do the 2 hits then either parry the knight, kick, back off, dodge to left or crouch at the right time.

    I don’t know all this 0.2 sec and windup time stuff but I just “feel” it, it felt like that knight was gonna get that hit in, also it does not look like latency issues.



  • @Hulking:

    Is everyone going to ignore how fast the first swing of the knight is compared to the second?

    Its really clear hes lagging and he just got desynced.

    He is standing nearer, like looking down overhead swing.



  • @Hulking:

    Is everyone going to ignore how fast the first swing of the knight is compared to the second?

    Its really clear hes lagging and he just got desynced.

    https://docs.google.com/spreadsheet/ccc … TZnc#gid=1

    The second swing was a combo. Combo windups are considerably longer, since you skip your recovery period to execute a combo. In the case of the grand mace slash, the combo swing is .225 seconds longer than a normal windup.

    @Elgar:

    @SlyGoat:

    Call him a bad player for slash spamming all you want, this could’ve happened to anyone - of course, it doesn’t happen to very good players specifically because they combo feint every single attack, as it’s the only way to avoid trading hits with the wonky flinch.

    So basically, a “good player” is someone who’s discovered a way to effectively circumvent a poor (“wonky”) game mechanic?

    Sounds more like exploiting weaknesses in the game than skilled playing to me.

    I don’t know about you, but I’d personally rather have a game requiring proper skills, than a game based on finding ways around inaccurate game-mechanics that result in the controls behaving in a unpredictable manner. In fact, a game claiming to be skill-based shouldn’t allow for these sort of things, or it would clearly not be skill-based in the true sense of the word.

    You misinterpreted the point I was trying to make. I stated that good players - those who are actually good at the game - always combo feint now, because it’s the only way to work around the inconsistency of flinching. The fact that combo feint is so prevalent to avoid hit-trading shows that flinch isn’t working as intended. Hence why I’m saying the OP isn’t necessarily a bad player for comboing - after all, combos are built into the game and meant to be used as he was using it, but because of the wonky flinch he died for it. (Granted he also mis-timed a parry and probably would’ve walked away with two kills if he hadn’t, but it wouldn’t have mattered if the flinch worked)



  • This bug happens frequently and needs to be fixed. I don’t get why there’s always that much discussion involved about this bug.



  • @Falc:

    This bug happens frequently and needs to be fixed. I don’t get why there’s always that much discussion involved about this bug.

    It is hard to tell due to all factors and not knowing exactly how it is suppose to work and the numbers behind it.

    I still request someone to make a video but just 2 people, little to no latency and reproduce it 10 times or something.



  • After seeing the videos and playing for some hours, i can definitely see that flinching is inconsistent and should be fixed.

    It’s also probably one of the reasons that 90% of players play vanguards and knights, because fast weapons that rely on flinch can be very inconsistent.

    I’m currently playing a lot with daggers, and i notice i cannot most of the time rely on my combos to flinch the opponent. I really just go in for a slash, parry, and only attempt to flinch the opponent if I have no other option, because half the time he ends up swinging his halberd anyway.


Log in to reply