A short time ago, at a computer terminal far, far, away, Pierre Perret wrote:
>willday@rom.oit.gatech.edu (Will Day) wrote:> >> Some _realize_ what the problem is, perhaps after playing similar games >> that do _not_ force-center, such as VWE Btech or Earthsiege. I've played >> these, and even then it took me a few games to figure out what was >> bothering me about the game controls. >> >> Most players, though, notice only that joystick targeting is very >> difficult. How many times have you heard someone say, "I don't use the >> joystick." or "I use the mouse for for targeting. It's much more >> precise." IS THERE REALLY ANY WONDER WHY??? > >Well, you are rather mistaken about the opinion of other players. Just >because you haven't heard much from people who are happy with the way the >joystick is currently handled in Mechwarrior doesn't mean they do not exist. >Remember that the disgruntled ones (i.e. your side of the fence) are always >the most vocal.
Well, I'm glad to finally meet one of you then. Have you, though, in fact played the other Mech-type games that do not force-center, such as VWE Btech or Earthsiege I? And, are you able to achieve as great a targeting precision with the centering controls of Mech2 as with the non-centered controls of the others?
>Frankly, I like forced centering for the turret movement. It makes a lot of >sense because the amount of turret swing is bounded, just as the tilt of >the joystick is. I find it very convenient to be able to get a tactile >feedback on how far I have swung the turret and also enjoy >near-instantaneous turret response. It helps a lot for subtle mech steering.
Realize, though, that we're talking about torso-turning, and not mech steering. Do you use the joystick x-axis for torso, or for chassis?
>To understand this better, just imagine having a head-tracking setup like >the Virtual-IO glasses or some such... If you mapped the turret movement to >the head tracking device, the equivalent of absolute mapping (a.k.a. >"forced-centering") would be to have your head position directly equate to >the turret's position,
Perhaps, but then that's a different input device entirely, so it may indeed work better with a different system.
One of the major differences, I think, is the size of the arc in which you can move the device. You have 180 degrees of movement in your head, from side to side, but there's only 45 degrees, at best, of movement in a joystick's x-axis. You may be able to move the joystick in 2-degree increments, but that means the torso will move in, at best, 8-degree increments. As a result, then, you lose massive precision when you map from a 45 degree controller to a 180 degree virtual movement range, and your aiming suffers drastically.
Of course, the length of the joystick handle would have an effect, as well, as that determines the real distance in which your hand moves. The farther you are from the center, the greater distance you have to move, and the greater precision you can achieve. Perhaps if we had a joystick with a 10 foot handle which we _walked_ from maximum to minimum, we could have the precision necessary to use this type of direct-mapped system. Or, if we perhaps had a joystick that had a full 180 degrees of movement, much like a head-mounted-tracker.
>whereas incremental mapping would have the amount you >swing your head dictating the speed at which the turret swings in the same >direction. I'd bet incremental mapping would wreck the immersion value of >such a virtual reality setup.
With a situation where you have a virtual range that's significantly greater than the input range, I think in this case that the incremental mapping _would_, in fact, be preferred. Imagine a game where you have full 360 virtual range, but you can only move the head tracker in a 90 degree arc. Not only would you have much less precision, but when you swung your head through the arc, the world would flash by you so fast you'd feel like a gnat on speed. Your stomach would probably like something else entirely, and I'm not sure that's the kind of immersion you really want.
Besides, when we're talking about a virtual device whose main purpose is targeting control, then precision is the preferred goal over immersion. I mean, you don't want to _feel like the torso_, you want to target _using_ the torso.
Also, since the mouse is the preferred targeting device for so many, imagine a mouse that's limited to only, say, 1 inch of horizontal movement, and that entire distance has to map to the 90 degrees of torso movement. You wouldn't be able to control the torso very precisely; your movements would be very gross, in large increments too big to pinpoint the body segments you wish to shoot, and you'd be forced to swing back and forth over the target in hopes that you can time it well enough to place a shot in the desired location. It would, I imagine, be very frustrating to you. Not to mention, of course, that the mouse would be made such that it always tries to snap back to the middle, so you're fighting the mouse at the same time as you're trying to pinpoint your target. That's what we have with the joystick, when it's directly mapped like this.
>I have a suggestion: you might want to try one of the joysticks out there >that have adjustable stiffness/recall. On some of them, you can tone down >the amount of recall to a point where the stick will stay in the position >you put it, with no re-centering of the handle whatsoever.
Yes, fighting the joystick springs is part of the problem. I still don't think, though, that it can achieve the precision necessary to target effectively. Besides, my Thrustmaster FLCS already has springs that are not particularly strong. I also resent the suggestion that I should replace a very nice $200 joystick, that works just fine in every other simulation I've played, just to get around a game that doesn't implement its joystick control in the accepted manner.
>Personally, after playing with MW2 and GBL for a good while (now beating >most MW2 missions on hard with a Jenner or, in a few cases, a Maddog mech >and having reached the trials of blood right a couple of times in GBL), I >don't feel hampered in anyway by my targeting on my MS Sidewinder 3D.
Sure, after awhile of playing MW2 and GBL, I got used to the system, dulled to its inadequacy, and I forgot just how much I'm missing. Recently, though, I've been playing Kesmai's Multi-player Battletech and Bioware/Interplay's Shattered Steel, which both use the incremental, non-direct mapping, as do, in fact, every other mech-type simulator on the market.
When I sat down with these two games for the first time, I was genuinely surprised at how immediately I noticed the difference - and I was quickly reminded of just how satisfying and exhilarating mech combat can be when you have exacting, precise controls. I could twist, turn, and backup all while flawlessly targeting a catapult jumping toward, over, and beyond me. Couple that with the fact that Mercenaries just came out, and did not include the option to change the torso-control system that we had expected, and that's why I'm so vocal again, as I was last year when Mech2 came out.
Sure, the incremental-mapping system may take more practice to use, and more effort to keep track of your current torso location in your mind, than MW2's simplistic direct-mapping controls, but the rewards are immeasurable in terms of flexibility and realism, not to mention accuracy, which _is_ measurably greater.
>I do >use the mouse for very long range sniping in a few rare occasions (like >shooting turrets from beyond their range while standing still and using a >large laser, or pummelling a dropship from 1200+ meters with a gauss rifle).
And, of course, you could do this without difficulty with the joystick, as well, if it were not using direct-mapping. Not to mention, you could achieve this level of precision even while moving your mech full-speed in the middle of a heated battle. _That's_ the kind of immersion I'm looking for.
===
Will Day