I notice practically no improvement, and little difference, in these areas. I see the default textures are different but those have always been replaceable and customisable in Arma. Whether they're "better" is purely taste (I personally don't like either game's defaults). And I see alpha blending is enabled by default (actually I can't remember off-hand if this is the case with Arma or not). Oh, and the font is a little different—rounder. I actually prefer the blockier Arma font (and have in the past suggested making it even blockier, actually). I notice no difference whatsoever in "smoothness."arilou wrote:From playing it for a couple of days it's far better than any Arma client (2.8 or 3.0 ) in the areas of looks (with or w/o movie pak) and smoothness.
On the subject of building up a new game from the foundation, I'm wondering what areas this will most noticeably include. First thing I think a lot of people would wonder about is rubber. That's a fairly (perhaps the) distinguishing aspect of Arma, however I believe that even Z-Man has lamented the lack of foresight on what would come of it, beginning as merely a compensation for latency (lag) but being abused to the point that it's now commonly accepted as a game mechanic or feature. And the rubber code in general, as I'm told, is a bloody mess. It's not explicitly mentioned in the WWWIA, but I wonder if that might be something that'll be addressed. Of course, I don't expect you all will eliminate it or its gameplay concept entirely—you apparently love 'HR,' the 'depth of grind,' and adjusting/180ing against walls, and all that orror. Unfortunately. But, still I wonder.
And if not that, what substantive changes do you have in mind?
[SIDEBAR]
If it were my project, I would eliminate rubber altogether. There would be something else to deal with latency, that would have no byproduct effect on the physics or gameplay. I would have a wall-touching system reminiscent of the game Tetris. You know how in Tetris, when your falling block makes contact with the others, you get a split-second to slide it before it gets locked in place. It would behave similarly, where you'd get a consistent, predetermined split-second to make contact with a wall before exploding. It could be configurable for being based on actual time or speed (distance traveled), but it would always be consistent. There would be no penalty for consecutive touches, as there is with Arma due to the rubber regeneration. And there would be no 'depth of grind.' Once contact is made, that's it. Only visible gaps could be passed through. Whether a cycle could pass through those gaps would be based on a configuration for whether the cycle was solid or not—or to put another way, itself affected by the lighttrace or not. If so, touching any part of the cycle model against a wall means contact. If not, one could imagine that these beams of energy only affect one another, not the cycle. However, if the cycle tries to pass through a beam wall, that will interrupt the wall he's leaving, breaking contact with the lightcycle and causing explosion. And, of course, there would be no ability to 180 or adjust against a wall you're already touching, with a configuration for distance that would be consistent and predictable.
I've actually tried to approximate this best I could in BiH and Shrunkland, but it's not there, and some of the behaviour is ridiculously, and irritatingly, inconsistent and unpredictable.