Jump to content
SubSpace Forum Network

Recommended Posts

Posted
Here is one reason why we can't have friction. If you slow down when you let go of the arrow keys, the client will have to send some kind of information to the server so that other players know they are slowing down. That sounds OK, but now imagine someone hitting the arrow keys, not holding them down, either loads of packets get sent to the server, or if there is a sample rate, people will see that player begin to slow down at a different place (far enough to dodge weapons).

This is ridiculous, plenty of games implement a drag coefficient, and so can cont. The major source of inconsistency in cont physics is the unpredictability WHETHER a player is thrusting or not combined with latency. Drag (as long as you dont allow it to vary per-player, which would be stupid) is completely predictable; there's no case (rather, you shouldn't allow there to be a case) where someone is not affected by drag.

 

If you added drag to cont, you'd have to be careful of a few things. First, cont probably internally makes some guesses about whether another player is thrusting based on pkt history. Those guesses would have to be reevaluated in the context of how players would behave if there were drag in a zone. Further, the total acceleration would be drag (always negative), plus thrust. If you start thrusting opposite the direction of your current velocity, the total negative acceleration will be higher than if it were just thrust. Since higher potential acceleration can cause greater inconsistency, zones would have to be aware of it and set 'tasteful' limits. As long as you didn't set the drag coefficient to Ragu's Thick and Hearty, this wouldn't be a problem.

 

-numpf

Posted

actually, there already is sort of a "drag engine", although there is no way to adjust it.

 

Pretend you're traveling in a straight-line. You gradually accelerate and start moving very fast. You then turn 90 degrees. You start flying "straight" in that direction (for a while, you actually move in a "slanted" path due to momentum from your past movement), and you start decelerating in your previous direction. After a while, your movement vector will no longer include the direction you were previously moving in, and you will travel straight in your new direction.

 

Although quite subtle, if you think about it, this is very awkward for conventional space zones (it's an explicit violation of Newton's 1st law, since no force causes that deceleration), yet in a ground-based infantry-type zone, having to compensate for past movement in different directions would seem ridiculous. So I personally think the drag engine currently takes on a "happy medium".

Posted
actually, there already is sort of a "drag engine", although there is no way to adjust it.

No there isn't. What you're talking about is truncation due to clamping speed to a max. That isn't, and shouldn't be called, drag. A 'natural' max speed is determined by drag vs max acceleration. We have an artificial max speed, which is necessary given the constraints that you don't want drag, and you don't want your physics to "explode."

 

-numpf

Posted
That is true, (notice how I only said "sort of a 'drag engine'" blum.gif ). My point was that this "truncation" system, although far from perfect, is something that already allows infantry-type zones to play less awkwardly.
Posted
My point was that this "truncation" system, although far from perfect, is something that already allows infantry-type zones to play less awkwardly.

No it's not. You have no idea what you're talking about, so stop. Infantry has some kind of drag, otherwise you wouldn't slow down if you stopped thrusting (or walking, whatever). How it's implemented exactly I don't know, but a max speed is different than drag.

Posted
I was comparing Continuum's speed truncation system to an ideal space physics system (namely compliance with Newton's 1st law), and pointing out how awkward applying space physics onto infantry-type zones would be. I wasn't comparing anything to infantry's drag at all. I know quite well what I'm talking about, though I admit that I made an ambiguous post -_-. And yes, it's pretty obvious that max. speed is different than drag. (though very similar since they both cause constant deceleration, albeit in different directions)
Posted

I think his point is that in an infantry zone, drag should be complete. When you are running, and then you stand still, you stop.

 

On the ground, you only move if you apply a force of movement. Thus, in a ground based zone, you should be stopped unless you are applying a force.

 

I think it really isn't worth it. Subspace is supposed to be a ship to ship based game, and ground warfare a distraction. I think ground warfare based zones should just say that they are fighting on a sheet of ice.

Posted

Qndre seems to be rather annoyed at all the poo-pooing.

 

It's a solid idea based on established physics approximations, and I'd love to see an implementation of these things in Continuum. Eh, mainly just for Metal Gear. ^_^

Posted

In real life, when you stop, you stop. You can do this using afterburners like in TW's tanks arena. Why not implement this in an infantry game? I see no need for a physics engine when it would be just as unrealistic as continous motion.

 

@catid: Check up on MGB forums, you'll see why there's a little poo-pooing going on.

Guest
This topic is now closed to further replies.
×
×
  • Create New...