Oolite Bulletins

Mad Hollander wrote:

Diziet Sma wrote:

(and if you're using Windows and are willing to spend $10, there is a truly brilliant configuration/profiler program available. [a free version is also available, but it's 4-5 years old now, lots has been added since then])

Is there such application in the world that can change the sensitivity of the joystick or set different sensitivities for different axises ?

Not that I'm aware of, no..

Though I might have to run Windows in a VirtualBox to see what the Saitek software is capable of these days.. I haven't run Windows in quite a while now.

Mad Hollander wrote:

Is there such application in the world that can change the sensitivity of the joystick or set different sensitivities for different axises ?

I've been searching and reading, and have an update to this.. some sticks do.. CH Products, for instance, allow you to adjust the sensitivity curves of the axes.. I also get the impression that at least some Logitech sticks do as well. Saitek promised it "this year" in 2007.. and are still yet to deliver.

It appears that such capabilities are generally left up to the games themselves to implement. Which is ridiculous, when you think about it, but there you have it.

Ideally, I'd love to see something like this in Oolite (doesn't have to be as fancy, of course, but just having the ability would be nice):

Ok.. found a possible solution for Windows users.. it depends if Oolite can see the vJoy virtual joystick or not, though.

http://theairtacticalassaultgroup.com/f ... php?t=6497

Looks nice, though..

Diziet Sma wrote:

On a scale of 0.0 to 1.0, I'd like it to be set at around 2.0 or even more. If it topped out at 5.0 or even 10.0 would be nice. As it stands now, I'd definitely still need to use precision mode in conjunction with this.

I set 1.0 and in precision mode I can fight on close and long ranges, but that's for my Adder. I haven't checked different ships.
Definitely, nonlinearity is good solution for both modes.

@ Mad Hollander: (or somebody) Would you be able/willing to check out the "possible solution" I posted immediately above your last reply? If the combination of software mentioned in that article allowed for programmable sensitivity curves for all Windows joysticks, that would be of use for lots of Oolite players.


@ cim:

cim wrote:

We do have this potential addition from kanthoney sitting around waiting for a decision - Github request #64

The way I interpret kanthoney's comment about a month ago, in the discussion at the above link

In the latest version, normal mode is linear and precision mode uses nonlinear with the value above (1.0 by default). I've also set it to save the precision mode setting between sessions - I can remove this if it's a pain. And I've added in a smoother transition out of the deadzone for normal mode.

would seem to suggest that non-linear mode only "activates" in precision mode.. is this the case in the way the code is currently implemented?

<wanders off to do some more testing>


Edit:

Yes, I can confirm that the non-linear effect is (currently, at least) only operational in precision mode.. where it works reasonably well..(I recall kanthoney mentioning that having a setting to make it optionally work in standard mode as well was a possibility)

Being able to set it about 1.5-2.0 times as high as the present maximum might be nice, but not essential, IMO. One thing I would like, though, to have is for the "knee" (transition-point back to linearity) set a bit higher, so as to allow for a little more stick movement before linearity kicks in.

@ cim, I've been doing a lot of reading on the subject today, and should you wish to tweak things further, there are a number of different formulae being used in various games for this purpose.. one of them may well be more suited than the current very basic one.

After a brief test, I don't think this is working too well. In fact, precision mode does not seem as effective now, even with the toggle set to zero.

Disclaimer: I know bugger all about the workings of joysticks - I only use them.

Strange.. works well for me.. I've found at setting "1.0" with extremely gentle pressure on the stick, I can get the sights to drift very slowly indeed.

Cody wrote:

In fact, precision mode does not seem as effective now, even with the toggle set to zero.

You're right.. and it explains why I managed to kill myself docking with precision mode when I first tried it.

The non-linear mode doesn't supplement precision mode.. it replaces it. Once you move the stick beyond the non-linear zone, you are suddenly in normal mode, i.e. NOT precision mode!

I checked this by coming to a halt mid-way between station and beacon, then timing a full-stick-deflection 360° pitch rotation, using the beacon as a marker. I got identical times in both modes, which did not used to be the case.. full-deflection in precision mode used to move at only half the standard speed.

In my opinion, this is not correct behaviour. Full stick deflection should still result in slower movement in the less-sensitive precision mode.

<nods> Thanks, Diz - you explained that nicely.

Okay - I can see what's happening here. Question: do you think that normal mode should be non-linear as well, or just precision mode? I can bring precision mode back down to the usual 1/3 turn rate either way, but since I can't test these changes it's easier to ask before I sent it to be built...

cim wrote:

Question: do you think that normal mode should be non-linear as well, or just precision mode?

<scratches head> My gut feeling is that normal mode should be non-linear too - but I'd only know for sure by testing it out.

<grins> Maybe I should ship my old back-up 'stick to you - none of the Devs use 'sticks, it seems.

Cody wrote:

none of the Devs use 'sticks, it seems.

Someone must, I'm sure, or how would the code have got there... but possibly no-one currently active.

I have considered getting one at times, but with 16,000 kills of flight time invested on keyboard control, it would be months if not years anyway before I could convincingly tell the difference between "this joystick change is a bad idea" and "this joystick change is fine, I just don't know how to fly this ship"

EDIT: non-linear normal mode, precision mode turn rate limits are now committed and should be in for tomorrow's flight evaluation tests.

cim wrote:

Cody wrote:

none of the Devs use 'sticks, it seems.

Someone must, I'm sure, or how would the code have got there... but possibly no-one currently active.

I think another_commander used a virtual 'stick of some description at one time - but I can't recall a Dev with a 'stick (mind you, that's only four years).

cim wrote:

EDIT: non-linear normal mode, precision mode turn rate limits are now committed and should be in for tomorrow's flight evaluation tests.

Cool.. looking forward to trying it out!

cim wrote:

I have considered getting one at times, but with 16,000 kills of flight time invested on keyboard control, it would be months if not years anyway before I could convincingly tell the difference between "this joystick change is a bad idea" and "this joystick change is fine, I just don't know how to fly this ship"

You might be surprised.. it's a very 'natural' way to fly.. you get the feel of it very quickly.


(And for the record, I'm inclined to agree with Cody, I think non-linear in both modes will work best.. but we ought to test it one step at a time)

Cody wrote:

<grins> Maybe I should ship my old back-up 'stick to you - none of the Devs use 'sticks, it seems.

It crossed my mind we should whip the hat around to buy them a couple..

Diziet Sma wrote:

It crossed my mind we should whip the hat around to buy them a couple..

Steady on! Old hardware is one thing, but actually buying them new kit? Hey cim... I've got an old ATI gfx card too, I think.