Jump to content
Sign in to follow this  
Cybes

Headsup for those unsatisfied with HOTAS/HOSAS

Recommended Posts

its a great idea and i applaud their efforts.  i hope they do well, regardless of whether or not my doubts are ill-founded.

 

my doubts?  many.  none of which were assuaged by watching their vids or the info on their website.  which only adds to my scepticism.

 

yes, i believe it is working as shown in the Descent reboot.  but thats deceptively forgiving, and not a good demo of precision or accuracy for all combinations of axes under independent or concurrent strain when rapid rates of change are required. 


its the nature of even 2-axis joysticks, that pulling a pure hard right, for example, without a skerrick of y-input is physically difficult for any mortal.  sure, unwanted inputs can be controlled somewhat with deadzones and response curves, but these arent cure-alls.  they must be adjusted for the specific demands of the task (game) at hand, to minimise a joystick's inherent weaknesses in each case. 

 

certain combinations are even harder.  try, for example, rapidly switching between NE (north east) at 80% of full extension, and ESE at 60% extension.  not only will you not be able to avoid overrunning or underrunning the mark, you will probably move the stick in erratic asymmetrical elipticals between those two points, rather than straight lines.  thats fine, you may say.  which is largely true, because we're very well adapted to this.  if you were controlling the nose of an aircraft you would simply make constant unconscious compensations for all those squirrelly accidental inputs without even thinking about it.  but if you tried doing the same with a joystick that has no mechanical damping, you would suddenly be very conscious of a huge proportion of your time and effort being expended to combat fishtailing.  so it goes, that a further requirement for precision is an ideal amount of mechanical resistance.  even with 2 axes, this is a delicate balancing act.

 

with the 3 axis joystick ive got, (z-axis left/right twist), i find it near impossible to fly a plane without accidental slipping and skidding.  thats partly, or even mostly, due to my lack of coordination and the quality of the joystick.  but the fact remains, the physical action is inherently imprecise when certain combinations of input are required.  in a dogfight, adding just a brief touch of rudder during pitching and banking is fundamentally difficult to do without introducing annoying accidental deviations to x/y input.  there's just hard limits to the level of fine control we can hope to squeeze out of one hand. 

 

ergo, imo the addition of translational movement poses major potential difficulties.  they have an interesting suspension system with force sensors at the joystick's centre of gravity, and granular measurement over very small movement ranges (1cm-ish).  pushing against that wire frame thing would seem to provide fine control with little effort, which is great. 

 

however, as it is, i cant imagine the resistance is easily customisable per axis, if at all.  going full-stick on all axes simultaneously may be overly resistive, for example.  and it looks like it would be very hard to give isolated input to one axis without smudging unwanted input across all other axes.  given the small movement range, taming said smudginess with deadzoning would seem very limited too, lest you want an axis that behaves like a near-digital hair-trigger that gives you RSI. 

 

it may seem like i am being defeatist.  i am not.  just saying, a lot of things need to come together for this thing to work as advertised. 

 

Share this post


Link to post
Share on other sites
4 hours ago, @~thehung said:

its a great idea and i applaud their efforts.  i hope they do well, regardless of whether or not my doubts are ill-founded.

...

it may seem like i am being defeatist.  i am not.  just saying, a lot of things need to come together for this thing to work as advertised.

 

You raise all the points I had as well, but... They /have/ been working on it for some time, and not just in Descent (heard about it through Star Citizen, and they mention that in the clip).

 

We'll see how it goes.

 

For me, that huge array of control buttons is bewildering, but I'd probably get used to it.

Share this post


Link to post
Share on other sites

yeah, i saw the SC clip.  and the drone one.  i hope they make something that addresses these concerns conclusively though. 

 

it seems pretty clear that at least for moderate rates of change and bandwidth of simultaneous input, and with moderate constraints on precision, it is highly capable.  that in itself may be more than enough for most contexts. 

 

then there is the possibility — which intrigues me — that even if there is an unavoidably high degree of crosstalk between axes, the sensitivity and the 'slew rate' of the sampling could each be high enough to render the necessary compensations intuitive and seamless.   ie. you just adapt unconsciously.  provided the extraneous effects were consistent right down to the smallest input levels, then that lefthand yaw you go for to line up a target feels to you exactly like a pure yaw movement, and the exact one you wanted, even though its a little bit smaller, because your brain has already factored in the highly dependable degree of "bonus left strafe" endemic to that amount of yaw 😄

 

 

Edited by @~thehung

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×