6 replies to this topic

[evilC]

Correct me if I am wrong, but it seems to me that you would need a very good quality stick to be able to use absolute aiming.

Using the JoyTester2 app, it appears my stick has a min-step of ~255
This would give it ~255 total possible positions for each axis (ie it is an 8-bit stick).

Assuming a mech with a 90deg torso twist, and playing in 1080p with a FOV of 90:

1920 x 2 = 3840 possible horizontal pixels that you would want to aim at.

3840 / 256 = 15x15 pixel block is the smallest possible granularity with this stick.

So I am guessing you would need a 12-bit stick (4096 steps) to make Absolute aiming viable?

With a 10-bit stick (Saitek X52) you would have 3840/1024 = 3.75 pixels is the minimum amount you can move.

With a 12-bit stick (eg BU8036 board) you would have 3840 / 4096 = sub-pixel accuracy (or you could play in >1080p with pixel accuracy)

Edited by evilC, 16 July 2014 - 05:58 AM.

[Winthrowe]

If you shift your perspective from 'we now have absolute aiming' to 'we now have positional aiming' in practice most of these issues can be fixed by an input curve. Few playstyles need full accuracy when they are fully rotated, since they're usually aiming at a close brawl target if they're that twisted.

[evilC]

I can see your point.
I tried running my normal stick (A TM Top Gun Afterburner II) through UJR (See sig) and adjusting the sensitivity.
At 50% sensitivity, you improve granularity by 50% for the first half of stick deflection, and decrease it by 50% for the second half.

This will not help muscle memory, and is the exact reason why any serious mouse aimer will disable pointer acceleration.

And it still only results in a min step of ~7 pixels for the sensitive center part, but will result in ~30 pixels per step for the deadened 2nd half (With an 8-bit joystick).

Edited by evilC, 16 July 2014 - 08:33 AM.

[shotokan5]

Do they want us to put in a cfg. file? I use an X52 Pro use their control settings and the game levers in the setting file.
Have had no problems using this for a while without file. Did I miss something?

[evilC]

shotokan5, on 16 July 2014 - 09:06 AM, said:

Did I miss something?

The changes are to allow "Absolute" aiming (stick half way right means twistED half way right not twistING right at half speed).

Until now you have been using "Relative" input.

[shotokan5]

evilC, on 16 July 2014 - 09:19 AM, said:

The changes are to allow "Absolute" aiming (stick half way right means twistED half way right not twistING right at half speed).

Until now you have been using "Relative" input.

Thanks for trying to give me an answer that I can understand. So it only really works in one
place under set conditions. Otherwise it is not working? Which means that PGI gave us a
joystick fix that does not work? Gee, how could that ever happen?

Guess I will run my joystick and throttle as I have which is ok.

This is more confusing than trying to pick a Canadian Beer in an Irish Pub.

[Loc Nar]

The resolution of the stick is important and low res will not help imake things better, but IMO this comes in behind the kinematics, which unfortunately can't be simply bought like you can with resolution cause no one makes a suitable joystick for positioning (well, right stick of a Steel Battalion controller...).

Bottom line is most sticks are mechanically way too sloppy when crossing the centers to even consider as positional controllers and fighting springs does not help this process at all. 16bits of precision in the Warthog (one of most mechanically challenged gimbals available) buys absolutely nothing in this game, whereas the 10bit res of my Cougar (even dumbed down with TARGET's mouse emulation) with an actual positional gimbal destroys it any day and I'd take the mechanical performance at 10bit of my Cougar over Warty's 16bit clunky mess any day (yes I own one).

tl;dr: resolution is indeed important, but without the mechanics facilitate positioning the point is fairly moot...