Firing Cone

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Tenshi

    #16
    Re: Firing Cone

    I just wanted to point out why it could appear that the bullets move in a cylinder when you look at the game engine.

    Comment

    • WarGimp

      #17
      Re: Firing Cone

      Well, I can't sleep, so I hacked this together:

      As you can see, the red area is always denser where you are aiming in the cone model. The cylindical model stays constant. Thus, it's actually more important to have good aim in the cone model if you want to make long shots.

      Edit: I changed the pic a bit to reflect range as a function of probability.

      Comment

      • evilzucchini

        #18
        Re: Firing Cone

        This exactly is one of the things that make me believe it really is a cylinder deviation it the game. When firing the PKM while standing for example, u can really miss most of the shots towards a guy that is standing just 1-2m away from you.
        When you are standing and moving with the PKM the cone becomes VERY wide. The crosshairs show you this. I swear that the bullets can leave the barrelll of the PKM at about 30 degrees and you are running around.

        Wargimp,

        Are you sure that the shape of the curve of the bullet deviation is bell curve? I haven't done any test to see, however, from my experience with the PKM for example (and nosekill has seen the same thing) is that the curve seems to be pretty flat; when my crosshairs are far apart with the PKM, I miss up close WAY more than I hit. I would expect with the bell curve to still hit more often than not. All i am saying is that I wouldn't be surprised if their was equality probability of hitting each point of any cross section of the cylindrical cone.

        Comment

        • WarGimp

          #19
          Re: Firing Cone

          Originally posted by evilzucchini
          Wargimp,

          Are you sure that the shape of the curve of the bullet deviation is bell curve? I haven't done any test to see, however, from my experience with the PKM for example (and nosekill has seen the same thing) is that the curve seems to be pretty flat; when my crosshairs are far apart with the PKM, I miss up close WAY more than I hit. I would expect with the bell curve to still hit more often than not. All i am saying is that I wouldn't be surprised if their was equality probability of hitting each point of any cross section of the cylindrical cone.
          Positive. Try the wall experiment.
          Probability math is a little tricky to explain... it's easier to show/see.

          Comment

          • Tenshi

            #20
            Re: Firing Cone

            Originally posted by Wargimp
            Well, I can't sleep, so I hacked this together:



            Edit: I changed the pic a bit to reflect range as a function of probability.
            That would mean some bullets (in the cylinder) would just stop in mid air.

            Comment

            • Tenshi

              #21
              Re: Firing Cone



              I wipped up a picture that shows what happens every time dx renders a frame i.e. how it 'adds' perspective to the world. The red cylinder shows the path of bullets that travel in parallel.

              Its a bit of a crude image but it should hopefully show how a conical firing pattern can be achieved without actually using a cone.

              Comment

              • WarGimp

                #22
                Re: Firing Cone

                Originally posted by Tenshi
                That would mean some bullets (in the cylinder) would just stop in mid air.
                No, it means that things farther away are smaller and harder to hit. Thus reducing probability.
                I understand what you mean about the apparent cyilinder, but since that's an artifact of perception, it doesn't interest me much.

                Comment

                • NoseKi||s
                  Member
                  • Jul 2006
                  • 481

                  #23
                  Re: Firing Cone

                  I agree with your wall example Wargimp and as i said earlier, when u shoot with for example a mounted MG it does look like the bullets spread in a cone.

                  Anyways.. i wasn't sure if they use a "gaussian" spread or linear spread (excuse my english) inside the cone/cylinder. I agree that changing from linear to bell curve spread changes the way your chances of hitting the target depend on your aiming and the distance to the target.

                  The biggest difference when changing from linear spread to a bell curve is that with a bell curve, a perfect aimer has the same chances of hitting regardless of the distance and the shape of the deviation in use. If it was a linear random spread and cone deviation, your chances of hitting would decrease depending on the targets distance no matter how well you aimed.

                  Still, even with the bell curve, the principle stys the same: with a cylinder deviation aiming well is more crucial than with a cone deviation. (just like u said Wargimp) The differences start to take effect as the aiming gets more inaccurate.

                  The reason why i argumented on this thread was the same post that the starter of this thread referred to. It was posted by some guy that claimed he had personally worked in the BF1942 team and stated as a fact that in that game at least they used a cylinder deviation.

                  From what i've heard it was possible to snipe with any gun as long as you just aimed carefully enough in BF1942, so maybe in that game thay still used a cylinder deviation.

                  Anyways.. Thanx Wagimp for an excellent example on how to reply to someone without flaming, even if u don't agree wih him 100% :)

                  Comment

                  Working...