Would you be interested in a custom HOTAS?

  • Definitely

    Votes: 12 31.6%
  • Probably

    Votes: 9 23.7%
  • Unlikely

    Votes: 8 21.1%
  • Not at all

    Votes: 6 15.8%
  • Yes, with caveats (specific features, OS, material, cost) (please list below)

    Votes: 2 5.3%
  • Other

    Votes: 1 2.6%

  • Total voters
    38
  • Poll closed .
I find myself disappointed in all the HOTAS available for Elite: Dangerous and Star Citizen. Their features cater to atmospheric flight, and are frequently unsuited to typical space flight navigation and combat. These issues are exasperated since I play from within an Oculus Rift.

I'm considering building a custom HOTAS setup specific for these genres.

edit: Adding rough design details.

Common Ground:
- No Buttons on the Base
- Mounting Options (Suction Cups, Screws, CH Feet, Weight)
- Adjustable Resistance on Sliders
- Adjustable to Hand Size
- Twist Axis with Rotation Lock
- Better 3-axis thrust control
- Gentle Center Detent for Throttle with Enable/Disable

Under Debate:
- Futuristic Styling, Blinky Lights (tm)
- Single USB Device vs Multiple USB Devices
- Force Sensing vs Traditional Joystick Motion vs Other (Air Mouse?)

Interesting Specific Features:
- Familiar button placements on Stick
- Analog hat on stick
- Warthog-esk extra index finger button and thumb hat
- One Hat and one Analog under Throttle thumb
- Clickable scroll wheel under Throttle middle finger
- Two buttons under Throttle index finger
- General Purpose toggle buttons within reach
- Device-Level Programmability (buttons, axis, deadzones, response curves)
- User Servicable
- Less bulky?
 
Last edited:
My current vision is loosely based on the Saitek X65, since Force Sensing plays nicely with Flight Assist Off mode. Personally, despite having my private license, I'm not a fan of pedals for gaming. Defaulting to twist for Yaw/Roll/whatever you prefer on that axis, but open to other suggestions.

Throttle/Thruster control feels like it could use a revamp as well. I'm thinking of something momentary, that returns to center when you aren't applying pressure, but with a friction lock on the throttle to keep it in place if desired... kinda like cruise control. Technically, we have thrusters for all axis, so it'd be nice to have an easy way to use them, preferably simultaneously. I'm using the CH Pro Throttle, and the analog stick is good for 2 axis, but not the third. Still thinking on this.

Which hand should house the UI controls? I've been using the throttle for now.

Thoughts?
 
So if I could design my dream HOTAS for Elite...

- A futuristic general feel, like the X52
- No button on the base of both the stick and throttle, since they defeat the purpose of not removing your hands from throttle and stick...
- Adjustable resistance for both stick and throttle
- One USB plug that splits in two cables for the two parts of the HOTAS, detected by Windows as one device
- Optional suction cups, and holes for permanent mounting with screws.

For the stick :
- Adjustable for all hands size
- General button placement like the X52, along with:
- An index button on the top right side of the stick, like on the Hotas X and high end sticks
- A thumb button like on the Warthog, Rhino, etc.
- Twist axis

For the throttle :
- Easy to access buttons like on the HOTAS X ; most HOTAS have awkward button placement that don't rest under your fingers and require you to change your position to reach.
- One hat and one analog stick under the thumb, and as many simple buttons as one can cram there.
- A clickable scroll wheel under the middle finger and two buttons under the index.
- NO buttons or anything under the ring finger and little fingers, these just need to rest comfortably.
- Don't know if this is mechanically possible, but a center detent for the throttle that you can easily disable when not needed.


I'd definitely like to see a new HOTAS, since none of the existing ones really suit my tastes ; I keep thinking "I like this one, but I wish it had this feature from that one"... I hope this will give you some ideas, good luck with your ambitious endeavor!
 
Please keep us posted on what you come up with.

3 axis thrusters is definitely a problem in need of an intuitive solution.

+1 and have some rep.

(I'd like to see any twist axis-lockable in case going to pedal later)
 
OK, here's my "wishlist" that would be based on something like the CH which looks way better than the X-52 I'm currently using (who needs those rotary things on the throttle?):

  • Each part presents itself as a combination of USB HID joystick, mouse and keyboard with a single additional endpoint for programming.
  • Each part comes with its own USB cable and complies with the USB standard for power consumption. A powered 4-port USB hub that has been tested with all devices comes in the box.
  • Programming software does only one thing: it pushes a mapping to the hardware that maps stick input to proper, primary joystick, mouse or keyboard events. No input event faking or custom drivers. (Hey, we're already the single most cross-platform stick on the market!) Ideally and in a pinch, one would be able to program the stick with a text editor and a very small CLI program without losing functionality. The protocol for programming will be fully documented and available to the public royalty-free.
  • Response curves including deadzones and ranges for axis inputs can be programmed into the device, again, no host-side mapping.
  • Industrial standard parts wherever possible. This includes microswitches, potentiometers, mini-joysticks, n-way toggle switches (which could then be easily swapped between permanent toggles and moment switches).
  • Adjustable tension and friction on all actions.
  • If the primary trigger has two stages, there will be the options to convert it to single stage by either blocking after the first pressure point, or by disabling the first pressure point and requiring pulling all the way through.
  • No arbitrary detents on the throttle. If there are idle or afterburner/whatever zones, they will be small, completely removable, can easily be mapped out of or into the movement range as needed, and come with extra input events that can be separately mapped to key events.
  • Easy access to all internal parts, no "warranty void" stickers, no springs that escape into the carpet as soon as you loosen the first screw, no cheap plastic threads that get stripped after the second time one drives a screw in them; a maintenance manual comes in the box and describes complete teardown and rebuild.
  • No lighting. Maybe a small display and minimal input to efficiently switch between profiles and debug inputs.
  • Stick rotation that can be properly locked without remaining play (I'm looking at you, X-52).
  • Bases as flat as possible with pre-drilled holes to screw them down, possibly with brass fittings. Also, with a built-in steel base to weigh the suckers down.
  • The rubber feet from the CH pedals. Accept no substitutes.
  • Comes in three sizes: Saitek, Human, and My Girly Little Hands :rolleyes:
 
Last edited:
OK, here's my "wishlist" that would be based on something like the CH which looks way better than the X-52 I'm currently using (who needs those rotary things on the throttle?):

  • Each part presents itself as a combination of USB HID joystick, mouse and keyboard with a single additional endpoint for programming.
  • Each part comes with its own USB cable and complies with the USB standard for power consumption. A powered 4-port USB hub that has been tested with all devices comes in the box.
  • Programming software does only one thing: it pushes a mapping to the hardware that maps stick input to proper, primary joystick, mouse or keyboard events. No input event faking or custom drivers. (Hey, we're already the single most cross-platform stick on the market!) Ideally and in a pinch, one would be able to program the stick with a text editor and a very small CLI program without losing functionality. The protocol for programming will be fully documented and available to the public royalty-free.
  • Response curves including deadzones and ranges for axis inputs can be programmed into the device, again, no host-side mapping.
  • Industrial standard parts wherever possible. This includes microswitches, potentiometers, mini-joysticks, n-way toggle switches (which could then be easily swapped between permanent toggles and moment switches).
  • Adjustable tension and friction on all actions.
  • If the primary trigger has two stages, there will be the options to convert it to single stage by either blocking after the first pressure point, or by disabling the first pressure point and requiring pulling all the way through.
  • No arbitrary detents on the throttle. If there are idle or afterburner/whatever zones, they will be small, completely removable, can easily be mapped out of or into the movement range as needed, and come with extra input events that can be separately mapped to key events.
  • Easy access to all internal parts, no "warranty void" stickers, no springs that escape into the carpet as soon as you loosen the first screw, no cheap plastic threads that get stripped after the second time one drives a screw in them; a maintenance manual comes in the box and describes complete teardown and rebuild.
  • No lighting. Maybe a small display and minimal input to efficiently switch between profiles and debug inputs.
  • Stick rotation that can be properly locked without remaining play (I'm looking at you, X-52).
  • Bases as flat as possible with pre-drilled holes to screw them down, possibly with brass fittings. Also, with a built-in steel base to weigh the suckers down.
  • The rubber feet from the CH pedals. Accept no substitutes.
  • Comes in three sizes: Saitek, Human, and My Girly Little Hands :rolleyes:

Nice.

Seeing as there's another thread on the subject ..
if it looks as close to exactly the same as the stick / throttle in ED the game as possible, that might be popular too.
 
Thanks for the feedback. I hit up the other thread as well, and consolidated the ideas so far. I'll eventually update the OP.

-----

Brief Summary

Common Ground:
- No Buttons on the Base
- Mounting Options (Suction Cups, Screws, CH Feet, Weight)
- Adjustable Resistance on Sliders
- Adjustable to Hand Size
- Twist Axis with Rotation Lock
- Better 3-axis thrust control
- Gentle Center Detent for Throttle with Enable/Disable

Under Debate:
- Futuristic Styling, Blinky Lights (tm)
- Single USB Device vs Multiple USB Devices
- Mostly Force Sensing

Interesting Specific Features:
- Familiar button placements on Stick
- Warthog-esk extra index finger button and thumb hat
- One Hat and one Analog under Throttle thumb
- Clickable scroll wheel under Throttle middle finger
- Two buttons under Throttle index finger
- Device-Level Programmability (buttons, axis, deadzones, response curves)
- User Servicable
- Less bulky?

-----

USB HID is assumed at this point, with the initial target being the Windows generic HID Gamepad driver. In general, every function will directly map to an axis or a button.

Device-level programmability only adds a little complexity, but mapping of buttons and axis is done through the in-game interface. Default values can be loaded for a specific game, but if you change them, you'll need to remap them in-game. I'll have to look into device-level axis remapping, but initially it doesn't seem too terrible (famous last words).

I have no issue publishing programming interface details.

One of the biggest complaints seems to be build quality. Glass-filled-poly and brass fittings are capital E expensive, unless you're building these units in the hundreds or thousands.

Full disclosure: my day job is new product development, specifically embedded electronics... but I'm just one dude, and this is a side project. Custom tooling wasn't in my initial scope. I was seeing this as something closer to a DIY with some professional support. All parts will be sourced from Digikey/Mouser/Arrow/Avnet/Newark/Future when possible. The controller PCB will likely be custom by necessity.

Speaking of scope... the other thread mentions panels and MFDs. That wasn't in my initial project scope either, but isn't out of the question. Touch screen wouldn't work from within Oculus unless it is texture keyed somehow, and/or in reach without releasing stick/throttle. It sounds like HoshiQaVam is already working on something similar.

The 6DOF videos were interesting. https://www.youtube.com/watch?v=dzHZrDXHiRk&list=UUc-DKEGeh4lQS1DXtqtiFZA

I recently spent a week hacking together an Air Mouse of sorts for AltCtrlGameJam. http://www.altctrlgamejam.com/games/5420b31e64ff4f2c5a10ccd5 It seems like the hand orientation mechanic could potentially be cannibalized.

Also, I've seen little comment on my desire to use mostly force-sensing on this design. I'm trying to reduce the need for moving systems. The throttle would be an exception, but does anyone have issue with force sensing for attitude and thruster control? I plan to build in some minor deflection.
 
Also, I've seen little comment on my desire to use mostly force-sensing on this design. I'm trying to reduce the need for moving systems. The throttle would be an exception, but does anyone have issue with force sensing for attitude and thruster control? I plan to build in some minor deflection.

You mean like the first attempt of a stick for the F-16 that was boycotted by pilots? ;)

IMO one of the major advantages of joysticks is that you have a lot of travel in your control which makes fine adjustments that much easier, and my intuition would be that most people can coordinate physical travel finer and more consistently than pressure. The moving parts are what makes sticks good.

Re the brass fittings, an alternative approach, at least for screws holding the case together, could be molded cavities to hold nuts; they don't even need to be shaped to hold them, just enough clearance to use a socket, and a hole for a conical head on the other side. That would remove the dreaded self-tapping screws in plastic in some critical locations and replacements are cheaply available from every hardware store (switching between metric and imperial is also pretty much a non-issue with some diameters). Probably over-engineering, but it would remove one point of failure.

(edit) while I'm here, a short rationale on the whole "multiple hardware HID devices with device-side programming" thing: it means you don't have to deal with input hooks, running any software as admin, custom drivers, none of that jazz. Assuming the programming is kept in non-volatile memory, once the device is programmed, it will fully work on every OS with HID support with no software requirements at all. You could even use it to control boot managers or your BIOS/EFI setup.
 
Last edited:
I find myself disappointed in all the HOTAS available for Elite: Dangerous and Star Citizen. Their features cater to atmospheric flight, and are frequently unsuited to typical space flight navigation and combat. These issues are exasperated since I play from within an Oculus Rift.

I'm considering building a custom HOTAS setup specific for these genres.



You missed an important option: Yes, but only one built by myself for myself, otherwise it isn't custom it's buying what they are selling.
 
Last edited:
It was my understanding that the F16 and F22 force sensing discomfort issues were resolved by adding a few millimeters of deflection. My current vision involves a set of bushings to allow for 1-2cm... tight enough to reduce lost accuracy due to return-to-center travel time, but wide enough to give tactile feedback and linear tracking.

That said, if everyone would prefer moving parts, I can look into it.

I had assumed the brass fittings were for corrosion resistance. If all we're worried about is making the device user serviceable, my first thought would be to countersink and epoxy some bolts into the enclosure. Alternatively, there are some flared rivet-style bolts that wouldn't even need the countersink or epoxy, and may be cheaper still.

Re: multiple devices... breaking it into separate modules makes it easier for me to develop, but more expensive as a cohesive whole. Mapping it to industry default HID protocols isn't hard, and was the direction I planned to go anyway. I don't like writing Windows drivers; I prefer staying in the embedded world.

Exposing a programming interface to remap things device-side is a bit larger of a task, and seems a little counter-intuitive if you can just remap the default HID buttons and axis in-game. Useful for axis sensitivity curves, yes, but is it really better to map that button to F on the device instead of in the game? Am I missing something?
 
Last edited:
You missed an important option: Yes, but only one built by myself for myself, otherwise it isn't custom it's buying what they are selling.
Interesting. I hadn't considered that people might be genuinely interested in purely custom. Since you would be building it yourself, is it really a valid option? That sounds like 'Not at all' to me.
 
Currently playing with a pair of Thrustmaster T16000M sticks with laterals(x left/right, y up,down, twist as for/aft like a bike throttle) in my left and and rotationals (roll/pitch/yaw as expected) in my right.

Still on the hunt for a good way to to throttle, as currently I take my hand off the left stick and use the small slider on the base. Possibly a "thumb" wheel up on the stick.

Pair of hats and 4 buttons on each stick is almost enough for combat.

I like buttons, pics of my custom centre "console" I am cobbling together.

qYAm3MUl.jpg
caJU3VSl.jpg
PvJjAWzl.jpg
 
Exposing a programming interface to remap things device-side is a bit larger of a task, and seems a little counter-intuitive if you can just remap the default HID buttons and axis in-game. Useful for axis sensitivity curves, yes, but is it really better to map that button to F on the device instead of in the game? Am I missing something?

It's just in case some actions can't be reassigned to different keys, or if you have something like a mouse equivalent on the stick (trackpoint?) you may want to put the actual mouse buttons in different places.
 
It's just in case some actions can't be reassigned to different keys, or if you have something like a mouse equivalent on the stick (trackpoint?) you may want to put the actual mouse buttons in different places.
Good point. Button swapping is the easiest programmability to implement, so no worries there. I got a basic axis-swapping prototype done last night, and it looks to be mostly manageable.
 
[*]Comes in three sizes: Saitek, Human, and My Girly Little Hands :rolleyes:
[/LIST]

Gets me to wondering how you plan to shape/ fabricate handgrip(s).
I guess if you model one version (and I'm imagining can be switched out, if held with maybe 4* allen head/ machine screws, one top and bottom on each of two split halves), 3D printing, of moulded grip section, gets you easy (easier) custom scaling on girth, done in 3D printer software. Buttons would remain in the same place though, on the internal stick chassis.

Though there seems to be (some) debate on how good the Thrustmaster Warthog really is; one thing I'd like to see in the end is a stick with a little physical weight. Not altogether the same thing as adjustable resistance. I'd like to get a little tired from using the stick. Not hollowed out and lightweight, reaching the station feeling like you appreciate a few minutes off the stick, to me adds realism. Obviously the HARDEST bush material you can manage, say if it's a ball and socket, a stick that's good for years.
 
Maybe for a bit of inspiration on a traditional, long travel stick: I've got an old Gravis stick that still moves rather smoothly after many years and a lot of abuse, comes with adjustable tension, and could easily be adapted to work with a range of encoders.

The design still strikes me as pretty excellent.

The base and mechanism (that stick uses a pair of pots driven by the cams you see on the top left and right of the cross):

Joystick Top by ngollan, on Flickr

Top view of the base:

Inner mechanics, top-side by ngollan, on Flickr

Adjustable tension:

Joystick Base by ngollan, on Flickr
 
Interesting ideas you have there.
I'm using a Saitek X52, it's a good flight stick but it would be better with some of the features you mentioned in the first post.
If you need any help with your project feel free to contact me. I can help you with designing the stick and creating 3D models for presentation and 3D printing.
 
Stick rotation that can be properly locked without remaining play (I'm looking at you, X-52).

This is exactly the sort of thing which I assume would be very hard to do, and why I voted 'unlikely'. Of course if I'm wrong then I would be interested - maybe industrial-grade components might be available?
 
This is exactly the sort of thing which I assume would be very hard to do, and why I voted 'unlikely'. Of course if I'm wrong then I would be interested - maybe industrial-grade components might be available?

From my crude understanding of engineering (note: that was just a wishlist, I am definitely very far from the process of actually building anything, I'm more of a programming kind of guy), putting a steel pin through a hole in the fitting for the rotation should do the trick. I'm not sure what Saitek did, but that mushy "lock" on the stick leaves at least a third of the total travel and could surely be improved a lot without breaking the bank.
 
Just some thoughts on the above, as they occur to me. I like a lot of what has been mentioned, but a few personal preferences, in the hope it may help:


Not a fan of momentary throttles, myself.
Most of what I've flown for real (which isn't that much) did not have momentary controls, primarily as that hand could often be needed elsewhere, if only for a moment.

Making the controls adjustable for hand dimensions is a nice idea, but learn from the attrocious job done on Saitek products and make sure they are of damn solid build quality. Something that feels great for the first week and then deteriorates to a rattley piece of junk thereafter is more of a hindrance than anything else... yes, my darling FLY-5, I'm looking at *you*!!
Also be sure that the design suits/adapts to handholds from both low down at knee level and high up on a desktop.

Pritty pritty lights are always welcome, particularly if they change when you switch modes, throw a switch or similar. Even better if you can choose the colour!
Be sure they are not too bright (FLY-5, I am again looking at you and your useless twin blue LEDs). Remember also to give them an OFF switch.

Adjustable resistance is always good, though scaling back to a level where it's almost imperceptible will be fantastic.

Twist Axis with Rotation Lock?
You mean you can disable the Yaw twist? Sounds like another useful option.

Which hand should house the UI controls?
Right hand, IMO.
I have the FLY-5, as you may have guessed, which uses a single 'nipple' rather than a fully-fledged hat. However, in flight this governs the strafing, switching modes puts it onto my freelook and in menus it governs the LRU&D movement. I find it especially useful in flight as pitching up while strafing down and similar movements work very and naturally. I'd prefer a 4-way hat as the main, with an 8-way to one side for the freelook.
As is, this one hat serves all my needs far better than multiple hats between stick and throttle - In fact, using the latter on a friend's X55 was such a chore as there were four hats needed for the same functions that my one nipple covers exceptionally well and it took a considerable effort to remember which one was assigned where!

Buttons on the throttle base - I rather like these, actually. Buttons, switches, toggles, all useful for the lesser-used functions like landing gear, ship lights and cargo scoop. I find them quite immersive, myself and they give you the excuse to put more pritty pritty lights on everything!! :D

Weighted - I like this, as I play on a megamat (big slippy mousemat thing that covers my whole desk) with nowhere to screw/suction things down.
Failing that, some kind of under-desk clamp like a monitor arm could be good. I could draw a picture, if it helps...

Thumb buttons on the stick would be good. One is fine, but two or three could be wicked. Same for the index buttons on the right side - Most HOTASes seem to feature one, but have room for two or three.

More buttons on the throttle, too!
Having at least one hat is probably a good idea (Power Management, I expect), but I find my left thumb likes buttons much better.

LCD Screen/Info Panel - Kinda sad perhaps, but it goes with the pritty pritty lights and opens the options for third party apps, etc. It also helps move the stick away from looking like it was specifically designed for XYZ flight sim and more like it belongs in a spaceship.

Force-sensing - Sorry, but no thanks. I like buttons on my phone keyboard rather than touchscreen and I like moving sticks rather than sensitive ones.


MOST IMPORTANT OF ALL: Options, options, options.
Some very good features available here, but wherever possible they should be removable/disablable/switch-offable. That will fit a far wider market than something built any one way over another.
 
Back
Top Bottom