Released: Multi-Throttle Support for IL-2

[MikkOwl]

Originally Posted by Ghost33

Hi MikkOwl, I have something wonky going on with my throttles sometimes, If I take one throttle while both are at 0% and bring it up it works fine but if I park it at 100% and then bring the other throttle up, sometimes (most, but not all) it takes that one parked at 100% and brings it back down to just bring it up again. its like both throttles on the g940 only run one in game, but if I move them together they then separate again and I get left right movement respectively.

Seems to only do this when both are at 0% or both at 100%

Very odd, and Ideas, suggestions?

Thanks, Ghost33

You might have left one of them bound to 'power' in IL-2 which could cause some wonkiness. I suspect this because to control more than one engine in IL-2 seperately, one must tell IL-2 to 'switch' to the engine one wants to move (if not selected), and then give it the power setting we want to set it to. Assuming the theory is correct, see if you can follow:

Device link is probably not as intensely updatingly fast as IL-2's in-game native controller assignments. When you first move up the one that is not bound to power in IL-2 to 100%, all works as intended. But when you try to move the other one (which IS bound to power in IL-2), it will be read first by IL-2, giving IL-2 an instruction to change the power reading to whatever position it is in - but because the 'last' moved engine is still selected (the one moved to 100%) it moves instead! And mere milliseconds later, my software reads the one you are moving as usual, and tells IL-2 to switch engine selection and to set the power to whatever, as it should. And multi-throttle, not knowing that the 100% engine was moved by something else, does not try to update that engine back into the proper position (it only sends updates when you move your controller into a new position, not otherwise).

Go check if the suspicion was correct. If so, just unbind it (try binding power to elevator trim or something, then bind that axis back to elevator trim, that clears 'power' from any binds) and you can stop reading the stuff below . If not the case, you must read on.

The software has not used any kind of special 'moving them synced' method since it made the switch from keyboard emulation to device link.

I want to try to replicate the issue but your description is a bit unclear to me.

1. Which version are you using of Multi-Throttle?

2. What kind of plane was it - single engined, twin, triple, quad?

3. How did you go about configuring your G940 throttles for Multi-Throttle? Did you UNBIND them both from 'power' in IL-2 as well?

4. The two throttles behave absolutely normal in every other situation for the same engine configuration aircraft?

5. Assuming you have hud log enabled, what does it say about which engines are being selected when moving a throttle and it malfunctions?

6. From what you describe, you put one throttle at 100% and leave it there. Then when moving the other, only the one you are not touching moves? What about the one you are moving by hand, does that one move at all?

[Ghost33]

Originally Posted by MikkOwl

1. Which version are you using of Multi-Throttle?

3.2

Originally Posted by MikkOwl

2. What kind of plane was it - single engined, twin, triple, quad?

Twin engine A-20DB-7 Boston MkII, 1940

Originally Posted by MikkOwl

3. How did you go about configuring your G940 throttles for Multi-Throttle? Did you UNBIND them both from 'power' in IL-2 as well?

The only thing I have bound in my IL-2 HOTAS section is Aileron, Elevator, Rudder, Brakes and Aileron Trim.

I used the
SETUP - G940 BASIC.exe
Optional - ENABLE R1 AND R2 AS TRIMS.exe

Originally Posted by MikkOwl

4. The two throttles behave absolutely normal in every other situation for the same engine configuration aircraft?

Did not try any other aircraft

Originally Posted by MikkOwl

5. Assuming you have hud log enabled, what does it say about which engines are being selected when moving a throttle and it malfunctions?

At the time I had the "left engine selected" "Right engine selected" disabled but I still had a % reading. The % went up and down with both g940 throttle levers but only one ingame throttle would move with confirmation on watching the engine gauges "rpm" as well

Originally Posted by MikkOwl

6. From what you describe, you put one throttle at 100% and leave it there. Then when moving the other, only the one you are not touching moves? What about the one you are moving by hand, does that one move at all?

I would take G940 Left and move it to 0%, ingame Left went to 0%. I would take G940 Right and move it towards 0% and ingame Left jumped to 100% then moved down to 0% with the movement of my Right G940 throttle


Its odd also because as I was writing this post I fired everything up again and now I cannot duplicate the behavior. It's as if nothing was ever wrong. very weird. one thing I did do was change in the .InI

"THROTTLE_1_REVERSAL_BUG_TWEAK_AMOUNT=1.546"

up from 1.508 to what it is above as my % still jumps like 51 53 55 56 58 60 58 56 54 52 51 50 and I was trying to clear it up a bit.

Ghost33

[MikkOwl]

Whatever it was (or wasn't) it appears to be gone and un-replicatable so who the heck knows.

That it moves in 1 and 2% increments up and down seems to be something about the math that IL-2 uses to read controls. One needs devices with massive amounts of resolution (if one changes sensitivity one can see that it moves in 1% consistently, on devices with already high resolution, but not otherwise). I suppose that the different numbers you see on the way back could be either the strange IL-2 math or that the reversal bug fix is slightly too much or too little, so you might be able to get it completely consistant with some of that tweaking.

When tuning the amounts originally myself when seeing if it could be done or not, I used the AutoHotKey (the programming language I made this stuff in) debug features where I could see the exact indicated positions of the axes, and then I would check how much one 'normal' step was, compared to how much the fix was 'off', then adjusting over and over until they matched almost exactly. But each throttle was a bit different, and other throttles can be too. Unfortunately for you there's no way to see the numbers straight, all you have to go on is the in game %, which could be unreliable.

Either way, changing that value by some fraction cannot be responsible. Even if set to crazy amounts it would not influence the other throttle in any way.

[Ghost33]

Thanks for the response and help provided. That is what I have been doing. Load up IL2 turn on your program and test, close your program, change the .ini and reload your program and test again. As far as I can tell 2% or the occasional 3% jump is the best I can get and quite frankly good enough for me and much better than stock IL2. Haven't had that odd Left Right throttle bug anymore either.

When you said you used AHK's Debug did you mean that JoyID identifier you provided with Joy1-4? I fired it up to look closer and I do get a full range of every number 0-100.

Ghost33

[MikkOwl]

I did not mean those, the JoyID thing does not even show decimals. It's just debug screens one can access if running the uncompiled "AutoHotKey" code with AutoHotKey installed, and it shows all the variables, their values etc. With that I could see that my left throttle would move 0.723 (out of 0-100) when moving forward as little as possible. When tuning the reversal bug amount I could see if, when reversing, it also moved 0.723 (it did not usually, it could be way off until I dialed it in closer and closer. But when pretty close, it might move 0,6 or 0,9 instead and there'd be no real percievable difference in IL-2. The amount in the .ini file is that amount x2 by the way, since it bugs in both directions).

And yep, full range in 1% increments with the JoyID program, and the accuracy is even higher than that (around 0.7 increments as mentioned above). In IL-2 it goes from 0% to 120% (122 steps) but the throttle has more steps than that available (about 140-150) yet IL-2 does not seem to read it properly, and similar with other more accurate devices so, who knows why. Something IL-2 related.

[Sokol1]

MikkOwl,

Is possible a version with "separated_engine_buttons" that support toggle switch, say: switch up, engine ON, switch down engine OFF.

[seperate_engine_buttons]
TOGGLE_ENGINE_1_KEY=1joy23
TOGGLE_ENGINE_2_KEY=1joy24
TOGGLE_ENGINE_3_KEY=no
TOGGLE_ENGINE_4_KEY=no
FEATHER_PROP_1_KEY=no
FEATHER_PROP_2_KEY=no
FEATHER_PROP_3_KEY=no
FEATHER_PROP_4_KEY=no

Cause:





Sokol1

[MikkOwl]

Originally Posted by Sokol1

MikkOwl,

Is possible a version with "separated_engine_buttons" that support toggle switch, say: switch up, engine ON, switch down engine OFF.

[seperate_engine_buttons]
TOGGLE_ENGINE_1_KEY=1joy23
TOGGLE_ENGINE_2_KEY=1joy24
TOGGLE_ENGINE_3_KEY=no
TOGGLE_ENGINE_4_KEY=no
FEATHER_PROP_1_KEY=no
FEATHER_PROP_2_KEY=no
FEATHER_PROP_3_KEY=no
FEATHER_PROP_4_KEY=no

Cause:





Sokol1

It is absolutely possible, in fact that is what I wrote for myself and used for a while (before releasing version 3.1). I will put this into version 3.3.

By the way, very nice switches you have there. I always wanted metal proper switches like that.

The engine on/off as seperate buttons work like this (up/down, choose yourself):

Switch up = Engine Start: selects the engine and presses 'toggle engine' function.

Switch down = Engine Stop: selects the engine and moves magnetos all the way to "off" and then up to "1+2" again. This shuts off the engine no matter what it is doing.

Note: when people use buttons, if it was only like this, the 'Engine Start' button would also turn off an engine that is running. To prevent this, the engine start button can only be pressed once - if the engine then does not start (engine restart in air etc), one must press the "Engine Stop" once, and then try the "Engine Start" again. This is not a problem for people like Sokol with two state switches, as it is not possible to press the switch twice - it moves between two states only - to press a switch again the switch must be moved to the 'off'.

This also means the program must be told if we spawn in the air (engine already on) or on the ground (engine off). Unless someone thinks of a better solution to avoid the 'engine start' shutting the engine off if pressed when engine is already running...

Hmm. Maybe I will leave it simple and allow the 'engine start' to also shut the engine off if pressed again. Switch people have no problems with this, and not many people without switches would want 3 buttons for each engine anyway to justify implementing a 'air/ground' extra button and the extra programming and documentation.

[Bearcat]

Originally Posted by MikkOwl

VERSION 3.2 FEATURES:

- Removed fuel mixture and radiator support (troublesome and not fool-proof; never heard of anyone using it either).

I can always add fuel mix and radiator back later if there's any demand for it. But for now, making all these things work in a more user friendly, auto-set up kind of way was complicated and time consuming enough - I was afraid I would simply not get this update out at all if I tried to include those features, due to being 'buried' with the amount things that needed to be updated.

I hope you can incorporate that into the next update.. That's a great feature... especially the radiator axis...

[MikkOwl]

Originally Posted by Bearcat

I hope you can incorporate that into the next update.. That's a great feature... especially the radiator axis...

I will put it back, fuel mix too. I mainly took them out before because it was a mammoth sized job to upgrade the whole thing to a more user friendly level (config free setup) and the documentation. I was afraid I would never finish that update if I had to do it for all features, so I prioritized.

Regarding radiator on axis (with bonus unecessary exposition! ): I found a Bf 110 manual from WW2 yesterday, written in English (it was based on captured specimens by the British who tested it and wrote a manual for their own pilots on how to use it). It had separate coolant radiator settings AND seperate oil cooler levers. In the game I noticed that when messing with the coolant radiator settings, the oil cooler levers actually moved separately. But like everything else, their setting copied themselves over to the other engine whenever selecting another. But hey, neat, separate radiators at least. Wonder if Team Daidalos will bother implementing the radiators separately or not (going to have to ask them). They do have radiator on axis, but their test vid showed the radiator levers moving in unison, as if not separate for each engine.


----

New features planned for version 3.3:

1. Radiator and fuel mix making a return
2. Possibility of prop pitch (because I have an itch for that).
3. Toe brakes (which have to manipulate the rudder to work properly on the ground) no longer manipulates rudder during take off, in flight, or landing.
4. Possibly .ini free setup method for individual engine buttons (toggle & feather).
5. Separate engine kill switches.

[Bearcat]

I actually have better simulated differential braking (better than stock) by changing a few lines in my User folder @ C:\Program Files\Ubisoft\IL-2 Sturmovik 1946\Users\wells. When you use rudder pedals and set your brake axis it will become either an X or a Y depending on which toe brake you press (L or R).. what ever it becomes.. just add the other line (If X add Y, if Y add X) and then make it a read only file and it wont change. I lie your idea better though...

[HotKey move]
AXE_RX JoystickDevice2=trimaileron
AXE_RX JoystickDevice1=trimrudder
AXE_RZ JoystickDevice2=rudder
AXE_U JoystickDevice1=-power
AXE_U JoystickDevice0=-pitch
AXE_V JoystickDevice1=-trimelevator
AXE_X JoystickDevice2=brakes
AXE_Y JoystickDevice2=brakes
AXE_X JoystickDevice0=aileron
AXE_Y JoystickDevice0=elevator