Sosna

Bump... I'm kinda eager to use this mod. Would not complain about a wip release.

I'd just like to say that this is my favorite sound mod out there. The only one I've kept for more than a few test sessions. Very consistent, uniform quality sounds - the main reason I stuck with the default sounds for so long. Anyways, keep up the good work.

Apparently it's already possible in ArmA 1 http://www.youtube.com/watch?v=1Cb9dSEznyg So, I kinda expect it now in 2

I'm curious Suma, what is the relation between x and y sensitivity? From my testing x seems to be at least 2 or 3 times as fast as y. Say I wanted equal sensitivity for both x and y, what value would I use for y if x=1.0? I see now that x is simply much much faster than y. I was unable to get a y increment from within ArmA's options - that' why I thought there was some other difference. I had to modify my .ArmAProfile file in order to set the y value high enough to notice the increment. Anyways, thanks for your work on the latest patch. The improved mouse movement has made ArmA much more playable for me.

This issue is not related to hardware, nor is it related to the Windows mouse settings. It only seems to affect the x-axis. Y-axis movement is perfectly smooth, with pixel or sub-pixel movement. The increment is not instantaneous. It moves smoothly, and is almost imperceptible - the problem is that it does not allow the player to aim exactly where desired. The player is forced between a certain number of pixels, the higher the sensitivity, the larger the increment. Very small horizontal mouse movements are ignored - until a certain point when the point of aim smoothly jumps the increment. To you guys who say you do not experience it, I would suggest testing again with the following conditions: -mouse sensitivity at maximum in ArmA options -no floating zone (makes it easier to notice) -prone (to reduce distracting sway) -M16, iron sights, fully zoomed in Aim at a distant/thin object (antenna, tree trunk, etc.) something no thicker than the m16's front post. Now make very gentle left-right movements (no wider than the front post), try to align the front post with the thin object(s). Compare horizontal movement with vertical. If you really don't experience the issue I'd like to see a video with the above conditions, because then it could be a hardware/OS issue. I highly highly doubt it though. I'm almost certain it's related to the fact that x movement is faster than y at equal values.

The easiest way to demonstrate it is to look through the iron sights of an M16 while fully zoomed in, and try to align the front post with a very slim or distant object. Cranking up the mouse sensitivity in the options will exaggerate the effect. With my regular settings the increment is almost equal to the width of the front sight post.

I can confirm that the problem is still there. Although it has been reduced to the point where I notice it only rarely. I play with both x and y sensitivity set in the middle. In the ArmaAProfile: mouseSensitivityX=1.100000; mouseSensitivityY=1.100000; It's very curious that it only affects x-axis movement... Even with y sensitivity set to maximum there is never an increment (at least not a noticeable one). This leads me to believe that x input is handled somehow differently than y input - and that this could be the root of the issue. I know that at equal values the x-axis is more sensitive than the y-axis. It's possible that this is caused by some value that is added to the x input (The value seems to scale with increased sensitivity). This additional value is also what causes the increment by my guess. If this is true, I would suggest removing the value, and just let the player set x sensitivity higher  than y in the options if needed. 1.12 introduced some kind of acceleration (pointer ballistics) so small movements are counted as slower, and thus the additional value is smaller and the increment less noticeable. However, I think if the issue above is addressed, it will remove the need for acceleration (Which a few have complained feels draggy), and x-movement will be as smooth as y. The only detriment being that players will have to readjust the x-y proportions to keep the old sensitivity feel.

As far as visuals go, in regards to the forest video in particular: LOD morphing is still very apparent. I find it very distracting, even at this low resolution. I realize though that the models need to change at some point... I would suggest making it so that LODs only transition while moving, and at high zoom levels (optics, binoculars). The models shouldn't morph when casually eye-zooming, or using iron sights. Also, the lighting in the forest shadows looks very bland - shading (aside from textures) seems almost nonexistant. HDR overbrightening dark areas may be partly responsible for the effect.

Nice work so far. If you guys can make it look more like the vbs2 abrams I would be very impressed. http://img246.imageshack.us/img246/9427/abramswoodland1fg6.jpg

Yes... Do want! The camera position has bothered me since day 1 of playing ArmA.

Smooth mouse movement is awesome. Thanks BI.

When I say flying in the context of this thread, I mean with helicopters. When flying planes input is centered with the mouse, and with joystick. Similar to driving. It should not be centering with the mouse, since long turns require excessive mouse movement - just like with driving. With helicopters both are non-centering. Ideal for mouse, terrible for joysticks.

Actually not. It´s the biggest nonsense they have served us. Go play the Delivery boy mission with mouse and keyboard and you´ll understand. It´s simply stupid and what I like most is that BIS does absolutely nothing against it while it´s one of the most wanted fixes since Arma has been released. I don´t take washy excuses anymore as they seem to have all the time in the world to embedd new features like VON as it will certainly be a new part of Arma 2 while we still have to wait for bugfixes that are so old that they are falling over their own beard. If this "control" system isn´t extended or fixed for Arma 2 I certainly won´t buy it. I have no interest in unguidable vehicles anymore. I simply have enough of it. I meant that when flying the mouse does not center. I agree though that mouse centering while driving is irritating. (Since it does center though, joystick driving does feel correct) It seems that right now we can have either: -The mouse behaves as a mouse, and joystick behaves as a mouse. (current flying in Arma) or, -The mouse behaves as a joystick, and joystick behaves as a joystick. (current driving in Arma) Obviously, we want the mouse to behave as a mouse, and the joystick as a joystick. I think the solution would lie in giving different devices their own input characteristics, rather than sharing input between devices, as I guess it is now.

I'm aware that the mouse centers when driving cars/tanks. However, it does not center when flying. That's fine for mouse control, but it also doesn't center for the joystick, when it should - That's the problem.

Thanks for your reply, Ondrej. I've read a bit about Pointer Ballistics, sounds promising. I hope it works out. Â

It would be really nice to have the explosion/gunfire sound effects vary at different distances If Arma 2 is to sound like a next-gen game, it will need a system like this.

I´m using a Logitech Marble Mouse and had this aiming problem since Arma has been released. Vertical it´s quite precise, but horizontal with ironsights it´s just not working right. When I try to aim at someone beyond 200 I have to readjust left and right to finally get my sights on the target precisely. Quite annoying and it got me killed more than I can count. I´d also be very pleased if BIS fixed the aimpoint. As it is right now it is just too big and hindering aiming when zoomed in. I guess there is a mod that fixes it but I´d prefer a unified BIS solution here. I use a Logitech G1 mouse. I think there's one common link here, IIRC you have a 7600GT AGP and so do I. Could it be in our setup? Video drivers or something? I'm pretty sure it's not a hardware issue. I use a Logitech MX310, and aiming is jumpy. With a Logitech Click! it's the same, and with an old Microsoft Optical mouse it's the same. If I remember correctly, it was even the same with my laptop's touchpad. As for video card, I have a 7950 GT, and my laptop has a Mobility x1400. On both systems horizontal aiming movement is jumpy. I think this is an issue that's apparent on all systems, but since it is very subtle, most don't consciously notice it. However subtle, it does make long distance shooting with irons very frustrating, and it's about time it's fixed.

In long distance shooting with ironsights the problem is most noticeable. It seems to go away with mouse sensitivity set all the way to the left, but that isn't really playable for most people. This problem doesn't occur in other games. Maybe if ArmA could dynamically lower the sensitivity when zooming with ironsights it would fix it? And/or when pressing the 'hold breath' button? I think this is already done for scopes based on zoom level, but I' not sure right now. Although it seems to be an issue with the way the game handles aiming. For me, vertical movement is perfectly smooth. Even with sensitivity set at maximum. Horizontal movement is incremental though, with the largest increment at maximum sensitivity, and the smallest increment at minimum (which is not noticeable). I'm guessing the problem lies in the fact that x mouse input is handled differently somehow than y mouse input. It's possible however that y input is also incremental, just that the increment is never large enough to be noticed within the whole range of sensitivity. (Just to note: It is specifically a mouse issue, I haven't ever experienced incremental movement with a joystick)

Actually... Go try the flying mode in Google Earth (no bull). The setup and such is bad, and it takes some practice, but this is basicly the way I want flying to be with in arma with mouse/kb. Ctrl+Alt+A in Google Earth (I think), else google it The benefit of this is that you have much greater precision with controls - Whether it'll be easier to hit a target is debatable though Where realism is concerned I think it's a better system. Google Earth does it nicely. Relative control for the mouse, (Cursor does not snap to the center like, for example, when driving in Arma) and absolute control for the joystick. "Cursor" is centered when joystick is. For another example, check out FlightGear It's free and open source. They also managed to have simultaneous relative input for the mouse, and absolute input for the joystick. It has a helicopter flight model quite similar to X-Plane's.

I typically use a joystick No... Currently, joystick movement in Arma is wrong. It behaves as if by tilting the joystick, you move the mouse cursor, and the cursor is left wherever you placed it once you center the stick. It is relative, when it should be absolute. Centering the joystick should behave as center input, but it behaves as no input. If it were absolute, centering the joystick would put the cursor in the center of the screen, holding the joystick tilted left would place the cursor left, and centering again would return the cursor to the center. I don't know if I can make it any clearer than that, it's not an easy thing to explain... With the mouse it's currently as good as it will get (relative input = relative movement). The other option is self-centering, similar to Arma's current driving input (Which most seem to find frustrating) Keyboard, however has options: Either it could be absolute, i.e. pressing no cyclic key assumes center input, and each cyclic key assumes input in that direction. (Typical in flight sims) Or, relative: where no key pressed assumes no input. It's Basically identical to moving the cursor around with the keys. Hope this helps.

Ah, I've figured it out. The problem is that Arma uses the same input method for the mouse and the joystick. As we all know, a mouse does not center as a joystick does. A non centering input makes sense in this case. The virtual joystick stays aligned with mouse movements. If the virtual joystick did center, mouse flying would be complicated. Unfortunately, when using a real joystick, the virtual joystick does not center itself. So the two joysticks are misaligned, causing the symptoms I have described. The devs may have done this to prevent input conflicts. For example: if a joystick was detected, but untouched, it would be constantly sending center input while the pilot tries to fly with the mouse. BI simply needs to keep the virtual joystick and the real one aligned. I'm not sure what that would entail technically, but it would make joystick helicopter flight handle like any other sim (flight model aside).

Missing the point again In X-Plane, releasing the stick makes the helicopter resume a default "attitude". In Arma, releasing the stick does absolutely nothing. This has nothing to do with flight dynamics, but how Arma treats control input. I apologize for any confusion - before I was describing symptoms, now I have realized the cause. If you visualize 2 joysticks: The real one in your hand, and Arma's virtual joystick. Arma's joystick does not center when released, while the real one does. This causes the situation where the pilot must move his real joystick in order to center Arma's virtual joystick. To simplify: the input is relative rather than absolute. It's most definitely an error on the developers part.

Horizontal mouse aiming movement is still incremental. Video demonstration here Note in the video it's not possible to align the sights with certain targets, which equals death in a combat situation. Also note that vertical movement is not incremental. I know I'm not the only one who notices this... Please fix this BI. I really dislike relying on luck in long distance engagements Â

Thats what happens when you bank in a helicopter. Think about it this way, positioning the cyclic 1 inch to the left will create more lift or the right half of the disc than the left. So as long as that cyclic displacement is to the left, it will continue to roll left. To stop the rolling moment you would need to have an equal amount of lift on the left and right half of the disc, so a right cyclic input would be needed until lift is equal on both sides. To stop the actual turn you would need to level the disc with the horizon by inputing even more right cyclic, now there would be more lift on the left side of the disc than the right. When the helicopter reaches a level attitude the cyclic would need to be again adjusted to a position where both sides of the disc are creating equal lift again. Ok... So every single helicopter sim has done it wrong - most of which had actual pilot consultation during developement, and some of which had been praised by actual pilots for their accuracy. I somehow doubt that. No offense to Arma devs, but I don't think they would pull off a more accurate flight model than dedicated flight sims - Nor do they need to, since Arma is a "Jack of all trades" game. However, what is needed is cyclic inputs that share basic consistency with other - more accurate - flight sims. A 5° bank (While moving at speed) in Arma will cause the helicopter to stay at 5° indefinitely and fly in a giant spiral forever. This is because Arma interprets the joystick/keys as having stayed in the 5° bank position, when the input was actually removed. This is the issue, and it has nothing to do with helicopter flight dynamics - which I'm well aware of. I've done my best to diagnose the problem with Arma's FM. You're free to do so also, rather than criticize my diagnosis without even fully understanding it.

It seems the point is being missed. It is not a physics issue as much as it is a controls issue. I don't know if I can put it any more simply: For Example: If you bank while in forward flight and release the stick, you must then pull the stick in the opposite direction to halt banking input. If you do not, the game is instructed to continue the bank input indefinitely. I can only judge from my experience with other sims. Since every one of which (including X-Plane) allows the cyclic to center, I am more inclined to trust their flight model than Arma's.