Cockpit switch & knob interactions

Hello,

Again a little test with those switches and knobs, to see if they're consistent with the usual DCS rules on mouse click and avoid the sad fate of the A-10C ;)

DCS 1.5.8.12162.404

Inverted mouse switch logic:

• Flaps power switch
  
• Compass lights/Test lights
  
• Armament mode control
  
• Fuzing control
  
• ... all 8 armament switches actually - the knobs are fine
  

Misc inconsistent behaviour:

• COMM1 has push animation, not COMM2
  
• MPCD push interaction is only 1-way for the corner, 2-way switches (DAY/NGT, ...)
  

Probably WIP?

• Stopwatch not interacting (Stopwatch start/stop)
  

Other/questions:

• INS mode know cannot turn to test from off, has to be turned clockwise all the way, on purpose?
  

Hello,

 

Again a little test with those switches and knobs, to see if they're consistent with the usual DCS rules on mouse click and avoid the sad fate of the A-10C ;)

 

DCS 1.5.8.12162.404

 

Inverted mouse switch logic:

• Flaps power switch
  
• Compass lights/Test lights
  
• Armament mode control
  
• Fuzing control
  
• ... all 8 armament switches actually - the knobs are fine
  

 

Misc inconsistent behaviour:

• COMM1 has push animation, not COMM2
  
• MPCD push interaction is only 1-way for the corner, 2-way switches (DAY/NGT, ...)
  

 

Probably WIP?

• Stopwatch not interacting (Stopwatch start/stop)
  

 

Other/questions:

• INS mode know cannot turn to test from off, has to be turned clockwise all the way, on purpose?
  

Funny, those bugs were found by the beta testers and fixed. I'll check again.

I'm going to necro this thread, because this has been bugging me.

Please, one logic, and one logic only, for mouse-clicking on switches.

2-position switches are currently any-click (left or right) toggles it, and this is bad. In VR it is not always easy to visually recognize, whether a switch is currently upward or downward, especially if you are viewing it from and angle; EG the switches to the right of the INS Mode Selector.

In Real-Life™ if you are not sure if a switch is in one position or the other, and you know that you want it in one specific position, you push it in that direction. If it wasn't already switched, it will now be switched --good! If it was already switched, you feel that it is already switched in the direction you wished --also good!

Most 3-position switches are left-click downwards, right-click upwards. However, most 3-position switches are downward = OFF or TEST, centered = AUTO, and upward ON. This is the exact opposite of the world-wide windows (not just MS Windows) logic of left-clicking something to activate it.

Ergonomically, all switches, even two way switches should work the same:

left-click = switch moves upward to next position, or if at the upper-most position = nothing.

right-click = switch moves downward to next position, or if at the lower-most position = nothing.

This should be the unyielding standard.

I'm going to necro this thread, because this has been bugging me.

 

Please, one logic, and one logic only, for mouse-clicking on switches.

You'll be glad to hear there is a standard! \o/

That's the image in attachment, which Derelor posted in answer to a similar report for the A-10C (which is beyond any hope, despite the patch I tried to provide and that I don't tire to remind them).

Razbam is usually one of the few developers, beside ED/Belsimtek, who follow this standard... though when it's WIP some knobs and buttons are sometimes acting up (or down) ;)

+1 on forcing right/left click for the 2-way switches, that would be more natural and much safer!

I would go a little further and even say that it would be good to have a key mapping standard for the common commands like starting/stopping the left/right engine, resetting the caution light, opening/closing the canopy, switching the lights, and so on. But that's probably more difficult because there should have been a set of common constants in the API for those and I don't think it's the case.

Thanks for your reply.

That is exactly what I described. The point is, that it has some large short-comings, for which there is no point. I believe I described clearly what merits my suggestion has. I'm hoping that it won't fall on deaf ears.

What worries me is a "that's good enough" attitude quickly leads to "good enough" becoming "easy-enough", which leads to many things being easy, but few being good.

With the additional VR argument to support this improvement, it should definitely be considered :) Perhaps you'll stand a better chance by posting also in the DCS wishlist (although I'm not sure, there must be tons of suggestions there...), or by asking Derelor how to proceed with this, he seems to know the subject pretty well.

This probably requires a modification of the default_button function (or similar name) for each aircraft, since the cockpit interaction logic seems duplicated for each module. I only know a tiny bit on the A-10C and the MiG-21 so I may be wrong.

In the mean time, I wouldn't be surprised if this could easily be done as a mod in the Lua scripts (typically cockpit/clickabledata.lua). And since those "in the mean time"'s can last for a very long time, the little problem is maintaining the mod with every update, I have done that for a while with the A-10C and the Fishbed then I finally gave up.

An argument I heard several times was "yeah, but if I integrate this, what about the existing missions, will that break them?", which of course is a false argument (at worst some mission may explicitly say "left click this or that" but it's very rare). The other resistance probably came from changing what the existing base was accustomed to, but then again, if that makes more sense they should like the change.

Anyway, I was talking about the general behaviour of all modules. Perhaps it's the wrong approach, and by trying with the Harrier and asking to Razbam if they'd take the patch... who knows. It's still very much in development and they might be more willing to improve things. If that proves to be better as a proof of concept (big words for such a small modification!), it could be easier to convince that the other modules should also be changed.

I'm going to necro this thread, because this has been bugging me.

 

Please, one logic, and one logic only, for mouse-clicking on switches.

 

2-position switches are currently any-click (left or right) toggles it, and this is bad. In VR it is not always easy to visually recognize, whether a switch is currently upward or downward, especially if you are viewing it from and angle; EG the switches to the right of the INS Mode Selector.

 

In Real-Life™ if you are not sure if a switch is in one position or the other, and you know that you want it in one specific position, you push it in that direction. If it wasn't already switched, it will now be switched --good! If it was already switched, you feel that it is already switched in the direction you wished --also good!

 

Most 3-position switches are left-click downwards, right-click upwards. However, most 3-position switches are downward = OFF or TEST, centered = AUTO, and upward ON. This is the exact opposite of the world-wide windows (not just MS Windows) logic of left-clicking something to activate it.

 

Ergonomically, all switches, even two way switches should work the same:

left-click = switch moves upward to next position, or if at the upper-most position = nothing.

right-click = switch moves downward to next position, or if at the lower-most position = nothing.

 

This should be the unyielding standard.

+1 this .

Funny, those bugs were found by the beta testers and fixed. I'll check again.

Quick check in DCS 2.5.3.24984

- "Flaps Power Switch" is reversed

- "Compass Light/Test Lights" is reversed

- "Fwd Equipment Bay ECS Switch" is reversed

- all ARMT 3-way switches are reversed ("Armament Mode Control", "Fuzing Control", ...)

- "HUD Display Mode Switch" is reversed

MPCD push interaction is only 1-way for the corner, 2-way switches (DAY/NGT, ...)

Stopwatch not interacting (Stopwatch start/stop)

So, yeah, basically all the bugs already reported one year ago are still there. I don't see them in the bugtracker, so I'm reviving that thread :)

+1 for having a consistent behaviour for all switches.

I add some more that act wrong :

• water injection
  
• nws
  
• fuel pumps
  
• apu
  
• gen
  
• eng start
  
• battery
  

Standard human interface for software is very simple, standard and part of the expected behaviour in any software.

If switch is vertical or is a 3 pos switch (e.g. Battery, APU, eng start, fuel pumps, nws, etc):

LMB moves switch UP, if switch already up then nothing happens.

RMB moves switch DOWN, if switch already down nothing happens.

If switch is lateral (e.g. Flaps power on/off switch)

LMB moves switch LEFT, if switch already left then nothing happens

RMB moves switch RIGHT, if switch already right then nothing happens

If switch is a rotary one (e.g. The INS align, nav, ifa knob)

LMB moves switch counter clockwise

RMB moves switch clockwise

I add some more that act wrong :

• water injection
  
• nws
  
• fuel pumps
  
• apu
  
• gen
  
• eng start
  
• battery
  

Those are correct, why do you say they act wrong?

+1 for having a consistent behaviour for all switches.

 

Standard human interface for software is very simple, standard and part of the expected behaviour in any software.

 

If switch is vertical or is a 3 pos switch (e.g. Battery, APU, eng start, fuel pumps, nws, etc):

LMB moves switch UP, if switch already up then nothing happens.

RMB moves switch DOWN, if switch already down nothing happens.

 

If switch is lateral (e.g. Flaps power on/off switch)

LMB moves switch LEFT, if switch already left then nothing happens

RMB moves switch RIGHT, if switch already right then nothing happens

 

If switch is a rotary one (e.g. The INS align, nav, ifa knob)

LMB moves switch counter clockwise

RMB moves switch clockwise

The standard is already in place (4th post), the only drawback is that for 2-way switches there is no check when the switch is already in the final position, it just toggles no matter what.

LMB is actually down, and RMB up.

Those are correct, why do you say they act wrong?

Because, e.g. to activate fuel pumps I need to right click. It should be left click.

LMB is actually down, and RMB up

no it's not. in software LMB means "activate", "turn on", or simply, "click to do stuff".

if you have a 3 pos switch where the upper one is "Start APU", the middle one is "Test", and the lower one is "OFF" then the expected behaviour is LMB to move from OFF to Test and again LMB to move to "Start APU".

no it's not. in software LMB means "activate", "turn on", or simply, "click to do stuff".

It is like that, did you look at the post I just pointed? You or I don't decide the standard here, it's ED ;) It is more important to have a consistent scheme than one that depends on whether a function "activates" or not a system, it'd be too confusing.

=> The switches you mentioned are fine, but the ones reported in the first post and a little above are still to be fixed.

and how come other modules respect the LMB -> UP and RMB -> DOWN.

it appears that someone can decide, or implement :)

Which modules? They all respect the standard given above by a tester, except the modules of Heatblur and M3L (the knobs turn the other way I think).

And except the A-10C which is completely inconsistent in most switches, probably due to the time it was built.

Please, this thread is to report a bug in the cockpit, not to start a debate. Feel free to create your own thread if you wish to do so.

Which modules? They all respect the standard given above by a tester, except the modules of Heatblur and M3L (the knobs turn the other way I think).

 

And except the A-10C which is completely inconsistent in most switches, probably due to the time it was built.

 

 

Please, this thread is to report a bug in the cockpit, not to start a debate. Feel free to create your own thread if you wish to do so.

No problem, I will leave the topic, the forum and the module to the bugs that suit you the most ;)

Quick check in DCS 2.5.3.24984

 

- "Flaps Power Switch" is reversed

- "Compass Light/Test Lights" is reversed

- "Fwd Equipment Bay ECS Switch" is reversed

- all ARMT 3-way switches are reversed ("Armament Mode Control", "Fuzing Control", ...)

- "HUD Display Mode Switch" is reversed

 

MPCD push interaction is only 1-way for the corner, 2-way switches (DAY/NGT, ...)

 

Stopwatch not interacting (Stopwatch start/stop)

 

 

So, yeah, basically all the bugs already reported one year ago are still there. I don't see them in the bugtracker, so I'm reviving that thread :)

Fix in attachment, tested in 2.5.4.25729.

Fix_AV8BNA_cockpit_mouse.2.5.4.25729.zip

Sent a PM to Razbam dev, no response whatsoever, I've uploaded the fix here for what it's worth:

https://www.digitalcombatsimulator.com/en/files/3303203/

I probably won't maintain it as DCS is updated, so expect that to become incompatible at some point and check before applying ;)