Check out the new MCOC AW "hit" show called 'How is this a hit?!?' ๐Ÿ˜…๐Ÿ˜‚ Featuring: Photon, & AbsMan.

TerminatrixTerminatrix Member Posts: 3,492 โ˜…โ˜…โ˜…โ˜…โ˜…
edited January 2024 in General Discussion
With the amount of
[S___P___A___C___E] between Photon and AbsMan's ball n' chain, there's no way possible she should have gotten hit. ๐Ÿ˜‚ I know the defender is supposed to be challenging, but c'mon now this is ridiculous! Can someone please explain?





https://youtu.be/zESii2drF10?si=YbPlvPJc_RIE0qTW

Comments

  • TerminatrixTerminatrix Member Posts: 3,492 โ˜…โ˜…โ˜…โ˜…โ˜…
    DrZola said:

    Cue up the usual โ€˜splainers who will tell you there is a *physics engine* in game that determines if a hit is a hit.

    Thatโ€™s a fancy coding way of saying โ€œjust listen to me and donโ€™t let your lying eyes deceive you.โ€

    Dr. Zola

    I just KNEW I was in the clear ๐Ÿ˜‚ Then *POW!* We're all of a sudden getting wrecked. ๐Ÿ˜… Wait...whaaaa...??
  • Wong_98Wong_98 Member Posts: 1,015 โ˜…โ˜…โ˜…โ˜…
    DrZola said:

    Cue up the usual โ€˜splainers who will tell you there is a *physics engine* in game that determines if a hit is a hit.

    Thatโ€™s a fancy coding way of saying โ€œjust listen to me and donโ€™t let your lying eyes deceive you.โ€

    Dr. Zola

    @DNA3000 wanna do this?
  • CrcrcrcCrcrcrc Member Posts: 7,980 โ˜…โ˜…โ˜…โ˜…โ˜…
    I believe they once said that the game doesnโ€™t operate off standard hitbox = contact rules for a fighting game, so itโ€™s based a lot more on internal code. Meaning however you choose to analyze it, a hit is determined by the range of it and what itโ€™s coded to do
  • DNA3000DNA3000 Member, Guardian Posts: 20,247 Guardian
    Crcrcrc said:

    I believe they once said that the game doesnโ€™t operate off standard hitbox = contact rules for a fighting game, so itโ€™s based a lot more on internal code. Meaning however you choose to analyze it, a hit is determined by the range of it and what itโ€™s coded to do

    https://docs.unity3d.com/Manual/CollidersOverview.html

    I presume, based on prior statements, that they are using mesh colliders. But separate from collision detection is collision processing, which gets into the MCOC specific functions.
  • Josh2507Josh2507 Member Posts: 748 โ˜…โ˜…โ˜…โ˜…
    The defenders seem to be able to hit us a lot easier than we can hit them sometimes

    Cassie lang (sp2 i think), a couple strikers, many champs heavy attacks. There's a lot of whiffing issues for attackers where the hit "visually" lands, but the attack doesn't.
  • DNA3000DNA3000 Member, Guardian Posts: 20,247 Guardian
    Josh2507 said:

    The defenders seem to be able to hit us a lot easier than we can hit them sometimes

    Cassie lang (sp2 i think), a couple strikers, many champs heavy attacks. There's a lot of whiffing issues for attackers where the hit "visually" lands, but the attack doesn't.

    Iโ€™ve seen a few like that, or rather I think I have, but they havenโ€™t happened at times when I was recording the fight to be able to analyze later. So I only have hunches and I think but I donโ€™t know kind of anecdotes.
  • ChovnerChovner Member Posts: 1,260 โ˜…โ˜…โ˜…โ˜…โ˜…
    Because... Hitbox reasons?!?!?! :neutral: lol

    We all know we're just waiting for 2 specific forum members to jump in tomorrow morning explaining how this is how it's supposed to work, and then hi-jacking the whole thread as usual.
  • TerminatrixTerminatrix Member Posts: 3,492 โ˜…โ˜…โ˜…โ˜…โ˜…
    edited January 2024
    Chovner said:

    Because... Hitbox reasons?!?!?! :neutral: lol

    We all know we're just waiting for 2 specific forum members to jump in tomorrow morning explaining how this is how it's supposed to work, and then hi-jacking the whole thread as usual.

    Lmfaoooo at least one is trying to make sense. I think the other one whose name has something to do with 'delusional wisdom', see that they have no winning argument here, and decided to sit this one out, which is where they should stay ๐Ÿ˜‚
  • TerminatrixTerminatrix Member Posts: 3,492 โ˜…โ˜…โ˜…โ˜…โ˜…
    edited January 2024
    DNA3000 said:

    DrZola said:

    Cue up the usual โ€˜splainers who will tell you there is a *physics engine* in game that determines if a hit is a hit.

    Thatโ€™s a fancy coding way of saying โ€œjust listen to me and donโ€™t let your lying eyes deceive you.โ€

    Dr. Zola

    Actually, it is possible physics had something to do with it, specifically physics on the ball. I went through the video frame by frame, and I noticed something.



    Here's the beginning of the swing. I'm going to show the very next frame just to show how fast it is moving, and how much motion there is between rendered frames (at least in the video):



    That's a lot of motion from frame to frame. Now look at how long the chain is. This is not a sprite, presumably, it is a 3D model. So it should be the same length throughout the animation, right?



    Okay, he's coming around for the second swing.



    There he goes...



    You can see, the game registered a hit, because it registered damage. But look at the ball. Is that where you think it should be? It seems to be a lot closer to AbsMan than it ought to be if it was actually swung out in front of him. Unless in between the previous frame and this frame (which in the video at least are a single frame apart) the game concluded that the ball struck Photon, bounced off, and ricochet forward.



    This is the next frame. Does that chain look distorted by the ball now? Like it is actually in an L-shape pointing towards the viewer?

    Here we see the flame from the hit on the ball. It is happening where the ball is, not where the contact point is, probably because that's how the VFX is programmed to work.



    Pretty obvious in the next frame that the VFX for the hit is out in the middle of nowhere, because where the target is and where the ball is have no correlation to each other.



    I've advanced several frames so we can see the next swing around. Note the length of the chain.



    This is literally the next frame. Between the previous frame and this frame, the ball has traversed from having not reached Photon yet to being swung past her completely. So there would not have been a visible video frame where the ball made contact. But let's check the distance. I'm going to put the second swing that looked like it whiffed and the third one where the length of the chain seems to at least be capable of having the ball hit Photon next to each other so we can compare the distances:



    Comparing the lengths of the chain, I think it seems like in the lower video frame, corresponding to the second swing above, the fact that the ball seems to miss by a mile is an optical illusion, a combination of the fact that the ball might not have taken a perfectly circular arc and might have struck Photon in between frames and the chain is now partially foreshortened. The upper picture shows that when the chain is in a different orientation there's plenty of length to reach the target at hit it.

    The problem in this case could be, as I've suggested previously, that the game engine might know a collision occurs even though the collision is never visible in a rendered frame. The game doesn't implement true motion blur, so we don't see objects sweep through each other. Instead we see them teleport around, making judging what happened with just our eyes tricky.

    Given the limited video evidence, it is difficult to be certain of anything. But it seems plausible that the problem is that the geometry of the ball and chain is long enough to hit the target, but it is moving so fast we don't see the actual collisions, and when the swing happens in the horizontal plane it can be difficult to see how much chain there actually is due to depth foreshortening.

    If someone has looked much more in-depth into the actual geometry going on here and has a different theory, I'd be happy to hear it. I can speculate, but I'm not a Unity expert.
    It's certain that she got "hit" when she should not have in the '1310' shot
  • SirGamesBondSirGamesBond Member Posts: 5,848 โ˜…โ˜…โ˜…โ˜…โ˜…

    Chovner said:

    Because... Hitbox reasons?!?!?! :neutral: lol

    We all know we're just waiting for 2 specific forum members to jump in tomorrow morning explaining how this is how it's supposed to work, and then hi-jacking the whole thread as usual.

    Lmfaoooo at least one is trying to make sense. I think the other one whose name has something to do with 'delusional wisdom', see that they have no winning argument here, and decided to sit this one out, which is where they should stay ๐Ÿ˜‚
    Chovner said:

    Because... Hitbox reasons?!?!?! :neutral: lol

    We all know we're just waiting for 2 specific forum members to jump in tomorrow morning explaining how this is how it's supposed to work, and then hi-jacking the whole thread as usual.

    Avoid provocative slanders.
  • DNA3000DNA3000 Member, Guardian Posts: 20,247 Guardian

    DNA3000 said:

    DrZola said:

    Cue up the usual โ€˜splainers who will tell you there is a *physics engine* in game that determines if a hit is a hit.

    Thatโ€™s a fancy coding way of saying โ€œjust listen to me and donโ€™t let your lying eyes deceive you.โ€

    Dr. Zola

    Actually, it is possible physics had something to do with it, specifically physics on the ball. I went through the video frame by frame, and I noticed something.



    Here's the beginning of the swing. I'm going to show the very next frame just to show how fast it is moving, and how much motion there is between rendered frames (at least in the video):



    That's a lot of motion from frame to frame. Now look at how long the chain is. This is not a sprite, presumably, it is a 3D model. So it should be the same length throughout the animation, right?



    Okay, he's coming around for the second swing.



    There he goes...



    You can see, the game registered a hit, because it registered damage. But look at the ball. Is that where you think it should be? It seems to be a lot closer to AbsMan than it ought to be if it was actually swung out in front of him. Unless in between the previous frame and this frame (which in the video at least are a single frame apart) the game concluded that the ball struck Photon, bounced off, and ricochet forward.



    This is the next frame. Does that chain look distorted by the ball now? Like it is actually in an L-shape pointing towards the viewer?

    Here we see the flame from the hit on the ball. It is happening where the ball is, not where the contact point is, probably because that's how the VFX is programmed to work.



    Pretty obvious in the next frame that the VFX for the hit is out in the middle of nowhere, because where the target is and where the ball is have no correlation to each other.



    I've advanced several frames so we can see the next swing around. Note the length of the chain.



    This is literally the next frame. Between the previous frame and this frame, the ball has traversed from having not reached Photon yet to being swung past her completely. So there would not have been a visible video frame where the ball made contact. But let's check the distance. I'm going to put the second swing that looked like it whiffed and the third one where the length of the chain seems to at least be capable of having the ball hit Photon next to each other so we can compare the distances:



    Comparing the lengths of the chain, I think it seems like in the lower video frame, corresponding to the second swing above, the fact that the ball seems to miss by a mile is an optical illusion, a combination of the fact that the ball might not have taken a perfectly circular arc and might have struck Photon in between frames and the chain is now partially foreshortened. The upper picture shows that when the chain is in a different orientation there's plenty of length to reach the target at hit it.

    The problem in this case could be, as I've suggested previously, that the game engine might know a collision occurs even though the collision is never visible in a rendered frame. The game doesn't implement true motion blur, so we don't see objects sweep through each other. Instead we see them teleport around, making judging what happened with just our eyes tricky.

    Given the limited video evidence, it is difficult to be certain of anything. But it seems plausible that the problem is that the geometry of the ball and chain is long enough to hit the target, but it is moving so fast we don't see the actual collisions, and when the swing happens in the horizontal plane it can be difficult to see how much chain there actually is due to depth foreshortening.

    If someone has looked much more in-depth into the actual geometry going on here and has a different theory, I'd be happy to hear it. I can speculate, but I'm not a Unity expert.
    It's certain that she got "hit" when she should not have in the '1310' shot
    You're certain. I'm not certain. I will leave it to the readers to decide how certain they should be.

    I analyzed the information provided in the good faith assumption that this was a discussion. That appears to have been an error on my part.
Sign In or Register to comment.