Technicly using vector coords, you can plot out pixel by pixel / quake unit by quake unit, each point that resides in the absmax/absmin values of the hitbox.
So you could map out all the points that the head area occupies and check versis trace_endpos if it encountered those points...but ah, the hitbox is not going to allow that right? True.
However, you could use some tricky Quake C to either resize the absmax of the hitbox so its neck height, then do your tracekine tests. Or possibly make the player a SOLID_TRIGGER perhaps. Would the traceline pass through that kind of solid? Traceline could let you ignore that entity if you had a way to reliably detect if its a client / player , but still you would neet to "step" the tracelines, sort of like the code thats out there that steps it backwards to detect if it hit water, then do a spalsh effect at the surface for the shotgun trace attack code.
I do know when you resize the hitbox to a non standard size, you have unreliable collisions, but if there should be a way to map out exact coords in a .mdl and match them to exact locations with respect to the model...and of course the frames, and possibly the angles might have to be taken into consideraiton.
So you could map out all the points that the head area occupies and check versis trace_endpos if it encountered those points...but ah, the hitbox is not going to allow that right? True.
However, you could use some tricky Quake C to either resize the absmax of the hitbox so its neck height, then do your tracekine tests. Or possibly make the player a SOLID_TRIGGER perhaps. Would the traceline pass through that kind of solid? Traceline could let you ignore that entity if you had a way to reliably detect if its a client / player , but still you would neet to "step" the tracelines, sort of like the code thats out there that steps it backwards to detect if it hit water, then do a spalsh effect at the surface for the shotgun trace attack code.
I do know when you resize the hitbox to a non standard size, you have unreliable collisions, but if there should be a way to map out exact coords in a .mdl and match them to exact locations with respect to the model...and of course the frames, and possibly the angles might have to be taken into consideraiton.
Originally posted by Dutch
View Post
Comment