Reticle depression axis

[scoobie]

Hi!
Have any of you tried to bind an axis to the "AN/ASG-31 Sight Reticle Depression Knob" (in control options for F-5)?
It doesn't seem to simply move the sight reticle up and down as, say, antenna elevation axis or whatever of this sort.
Instead, it regulates the speed at which the reticle is travelling up or down the glass, where the axis's very center is "speed 0", reticle stopped.

Move the axis a bit off the center - the reticle starts traveling, move it even furhter - the reticle speeds up.
It's essentially useless like that

Quite original, but useless.

Feels like a "minigame" inside F-5

 

Yes, I know that the in-cockpit knob makes 3.5 turns whereas a typical "home potentiometer" (which I use to control pipper depression in all A/C which have some kind of a knob for it) travels by approx. 0.75 of a turn only. Nevertheless, a pot could give enough precision, it's only a question of whether it's "quiet" enough, how big a knob you put on it and besides, there are multi-turn pots out there if somebody wants to use one (I don't). For the sake of retaining the best possible resolution the axis at the center could preferably translate into the reticle's center of the whole travel range, and not depression value of 0, i.e. the reticle goes from 35 up to 200 down = 235, so axis centered could give depression of 117 or so.

 

No, grabbing the mouse with your third hand, clicking the knob with it and now sweeping with the mouse up and down is NOT a better solution... simply because it's a worse one if you have knobs/swithes/etc. on or next to your throttle (I don't use VR).

 

[LeCuvier]

Yes it's weird in the way it works, and too fast. That's also true if you use it with a pair of pushbuttons instead of an axis input.

I find that it gets more manageable (slower) if I reduce the gain value in "clickabledata.lua".

The original line looks like this:

elements["PTR-LVR-SGT-RET-DPR-42"]    = default_axis(_("AN/ASG-31 Sight Reticle Depression Knob - Rotate to adjust manual mode depression angle"),  devices.AN_ASG31, asg31_commands.RetDepression, 42, 0.0, 0.1, false, true, true)

I reduced the last numerical value from 0.1 to 0.02 and that made it more usable, both for the axis and for the pushbutton commands.

We shouldn't have to mess with these files; but I have had to make many changes like that over the years because ED won't make them.

[scoobie]

Thank you, LeCuvier! I will do it like this. Yes, since the analog axis is virtually unusable, I've been using +/- switches for reticle depression already and, as you say, it IS annoyingly fast (like quite a number of other controls on various planes).

 

After a little of reflection, I think ED might have done this axis this way as a "self defence" measure. People may have less than perfect pots and 8-bit axes and if such hardware didn't work for them, they would rant about it, while ED can't do miracles. The full depression range is 235 values, so for 8-bit axis even a slightest noise/tremble (or even the A/D converter's inaccuracies, quantization error, all other errors) will make the depression readout drum tremble and people would go crazy. And these values aren't even discreet, the drum is smoothly turning from one value to another, e.g. from 134 to 135, so you can have 134.3 for example. Yeah... I think that was the reason they did it like this.

 

Still, it's a bit of a pity for me, I understand the hardware limitations, so I wouldn't rant  If 8-bit axis proved insufficient, then I'd just look for a controller with axis/axes with more resolution and perhaps a better pot or a multi-turn pot.

 

I guess I must invest in rotary encoders at last (though I don't know if they work well in DCS). I imagine this is how depression knob should be used. Leo Bodnar has nice double encoders with push, but they're expensive and I'd have to destroy my button box and build a new one... and I'm so lazy. Nevermind

 

Thanks again for the recipe, I'll give it a go today

 

[LeCuvier]

56 minutes ago, scoobie said:

...I guess I must invest in rotary encoders at last (though I don't know if they work well in DCS)...

Yes, rotary encoders usually work well where the pushbutton commands are too fast. But sometimes they are too slow. The dilemma for ED is in these cases where we use digital inputs for rotating knobs, that they really would need to provide two versions for each command; one for pushbuttons and one for rotary encoders. That would cause a big number of additional commands and might be confusing.

A better solution in my opinion: Add an option for "digital axis" commands, that allows you to switch between high and low gain. Or even to select the numeric gain value.

Edited February 11, 2021 by LeCuvier

[scoobie]

Thanks for the tip about rotaries!
Yeah, sometimes I feel so sorry for ED, they're developing HUGE software, and even such a "tiny" side issue (seemingly!) as those control assignments are such PITA, people are using different controllers, if they use up/down buttons to control some analog value (axis), then either it's too slow and takes ages to go from one extreme to the other, or if you make the buttons work faster then it's too twitchy/imprecise etc.
What a headache!

Some people here have suggested that "variable speed" control option could be the answer - for button pairs, not rotaries. I think... P-51 (?) has this feature for something - you press shortly and get a minute increse of a value, press and hold and it starts accellerating.
For rotaries, though, such "digital axis" with adjustable "gain" sound like a good answer!
But I can see ED devs jumping out the windows in panic

 

I've just got another idea for the depression knob for the Tiger. First I'll try your settings (for an (ON)-OFF-(ON) toggle switch I use for that), but then this idea: I could use the "mouse nipple" on TM Hawg throttle. I'll set some dead zone and then put a lot of curvature on the axis, thus creating something like a "two speed switch" - pull the nipple some - slow speed (precision), pull hard - FAST. Maybe this will feel good. And MAYBE that's what ED had in mind? To use it with a minijoystick? So far I've tried depression on a classic potentiometer and it felt ugly, meh, no-no. If the nipple works nice, I'll post my settings here.

 

----
(Off topic.)
So far I've refrained from builing a new button box. My biggest trouble is that I have limited space (I have to keep all mu stuff on the desk) and planes in DCS are just so different. I need to somehow "average" external lighting panel so that it's suitable for all (or most) planes, then average controls for radios, weapons (master arm, laser, A/A, A/G etc.) and all that stuff. It's so hard to find a sweet spot - a well balanced "average" set of controls, but at the same time a "slim" one (instead of five quare feet of buttons and knobs).
Anyway, rotaty encoders are extremely appealing here - using a surface area on the box's front plate of a (very) big button, a double rotary with push gives you two "axes" (two sets of +/- switches) plus a push button. The only trouble is that Leo is charging around $35 for one (they're most likely quality stuff though). Multiply it by 6 or 8 pieces (if not more) and you get a hefty price already - for the rotaries alone!

And at the end of the day... what if one day I'll try VR and - like so many others - I'll never want to return to "pancake simming" again? How do they deal with controlling stuff? Will my meticulously designed "super button box" become useless?
Oh boy, flight simming has always been such trouble... perhaps I'd be better off buying a stretch of land and growing carrots instead

 

[scoobie]

OK, I hereby testify that both solutions work nice
With 0.02 gain as laid out above by LeCuvier (and a pair of buttons), the reticle is perhaps a bit slow, but lets you easily set depression as you like.

 

The minijoystick works well, too. Gives more flexibility than buttons, but the downside is that minijoysticks are "expensive" (you don't have as many of them as plain buttons/hats, typically).
Anyways, my initial settings:
Dead zone: 5.
Saturation X: 100.
Saturation Y: 100.
Curvature: 60.
Tweak to your liking. Think of it as "two-speed buttons" (like 6-speed gearbox on a car). Hence the extreme curvature.

 

If you use your minijoystick primarily for Fire Control Radar TDC, you may use "modes" (in Joystick Gremlin or whatever you like to use) - you may assign the minijoystick's button ("click") as the mode switch button. In one mode you have FCR TDC cursor, in the other - reticle depression control. You don't do A/A and A/G at the same time, I think, so such modality shouldn't pose a problem.

 

Edited February 11, 2021 by scoobie

[LeCuvier]

18 hours ago, scoobie said:

With 0.02 gain as laid out above by LeCuvier (and a pair of buttons), the reticle is perhaps a bit slow, but lets you easily set depression as you like.

You can vary the gain value to suit your preference. So if 0.02 is a bit slow, maybe 0.03 is better