I may not have time to address all the concerns raised right now, but here goes.
1. I do hope to make the battlescape draw on game mechanics at play in many tactical-sims, within the bounds of a turn-based game like this. Fire and movement, cover and concealment, overlapping fields of fire, bounding overwatch -- these are all small unit military tactics which have strategic relevance in a tactical sim and which I hope to make more relevant in the battlescape gameplay over the next several years. We're missing some major components -- a visibility system, for instance -- but quite a bit is already in place: overlapping fields of fire, bounding overwatch and, via the TU system, some aspects of fire and movement.
UFO:AI is, of course, not a tactical sim. But these are critical aspects of a truly strategic 3D environment. They don't matter for the games-as-film genre of FPS, like Modern Warfare, or the arcade genre of 3D games, like Quake and its successors. But in my opinion they are an important step beyond the number-crunching world of turn-based tile combat, like Battle for Wesnoth, and a positive move towards a more interesting 3D strategic environment. It is a very early work in progress.
This is also why I have sought to develop mechanisms for tying weaponry to more specific combat roles. The sniper rifle is designed for a very specific purpose: to deliver a strong punch at a great distance. In return, it sacrifices mobility and flexibility. I have made its snap shot less useful because that is a firemode it is not designed to do well. It should be an option of last resort. That said, I may reduce its TU to 12, as only 5 less than the aimed shot seems wrong and was perhaps an oversight on my part. For more info on combat roles, read the
Skills/Weapons section here.
2. I have heard from former armed services members the same as Kurja mentioned: a crouched position is awkward for firing a machine gun. That's why I chose to put the crouched penalty in. If this is patently false, I will gladly reconsider.
3. I am against a simple equation between crouch and better accuracy. Currently, our maps really lack a lot of good cover -- objects that provide defensive firing positions. This reduces the element of cover in the cover and concealment game mechanics. But I hope in the future with more cover in maps, that stance will have more to do with the demands of a particular defensive position than just accuracy. The decision to crouch or stand could have implications on whether the player is more or less exposed and, as a trade-off, is more or less capable of returning fire. In my ideal future UFO:AI, this is a more important calculation than weapon accuracy.
4. On the big, heavy, unweildy element of sniper rifle snap shot accuracy: consider not just its weight or size, but also the fact that a sniper rifle is typically mounted with a large scope which improves the ability of its operator to accurately aim at long distances, but obscures the operator's view considerably, making it more difficult to visually locate and center the barrel on a target quickly (unless the operator is already dialed into an area). Now, our current models don't all visually show this, but that is something that I hope to improve in the future.