I think I've said that. In computer first-person-shooting games, you turn = you turn WHOLE BODY. Even though INF has a unique feather called "free aim area," the mouse sensitivity is THE SAME in this area.
In real life you don't turn WHOLE BODY to aim. Especially when you concentrate on an accurate aim, probably for a long-range shot, your hands can do very minute adjustments to your aim.
Another thing I want to argue -- a shooting position does not affect your physical aiming resolution in real life. A shooting position in fact should affect your steadiness and difficulty to aim, not resolution.
Currently we obviously have some problem with resolution. We know it is because of hardware limitation. I have figured out a way to represent a "real aim" even better than the current way and also avoid the hardware resolution problem.
I will write my idea down in the following paragraghs, but it seems to require a total re-write of RA, which I doubt yurch is willing to do.
P.S. I believe the resolution problem will become even bigger and more important in the future when the scope visual effect is implemented right.
---
The current problem is that the program requires our pointing devices ( mice ) to work at a higher resolutin than the hardware's physical resolution. I assume 300 dpi as the hardware's physical resolution here. ( This is a very stupid and tragic limitation if your OS is Windows 2000. Windows 2000 limits the mouse resolution at 300 dpi no matter how high the resolution your mouse really has. I don't know if WindowXP does the same stupid thing, though. ) Even if you have an optical mouse, it can only give 800 dpi and no more. That's just not enough to give a perfect scroll without causing the grid-effect when there is a 12x magnification, and it's even worse if your screen resolution is higher.
What if we make RA work under 300 dpi but still keep the same effect as it is now? Is that possible?
I think yes, if we include this idea proposed by many other people -- weapon inertia.
When a player starts to turn, RA makes the turn accelerate and decelerate. During the acceleration and deceleration, keeps the hardware request to 300 dpi ( if it is more than 300 dpi ), and there is a state between the acceleration and the deceleration, in which the hardware request works in exactly the same way as the current RA -- keep the same angular velocity. If this is done right, there will be no grid effect for very careful aiming, and if a player wants to make a hasty aim ( he/she moves his/her mouse very fast ), firstly the inertia will make it difficult to obtain a good aim; secondly, RA can be made as such that if a fast mouse movement is detected, cancel the player's concentration.
Also I would like to suggest another concept be implemented -- muscle tremble for standing and crouching shooting position. This is an angular effect, which means the more magnifying power, the more obviously this effect will be seen. And of course it is bigger when standing than crouching.
Undoubtedly, what mentioned above needs some tweak to get a satisfactory set of coefficients used by the whole process. But it is more REAL.