MiG-29 Fulcrum DCS

Good afternoon,
Thank you very much for your feedback.

I haven’t yet tested the C SDK for Command 604. I’ll try that next.

Thank you also for replicating the behaviour of the C++ SDK. We remain particularly interested in MainCtrl_FfbConstant (604) with enableAxis = X:0, Y:1, DirectionEnable:1, as this exact configuration works in the ApiDebugTool but not via the generated C++ request.

I will get back to you with the results of the C-SDK test.

Kind regards,
Matthias Rausch

Subject: Problem with Flight SDK – MainCtrl_FfbConstant (Command 604) on the AB9

Dear MOZA Support,

Thank you very much for your advice to use the MOZA Flight SDK in conjunction with the MOZA Device Service for aircraft-specific force feedback effects.

I am currently trying to implement a simple additional force feedback effect for the DCS MiG-29A on my MOZA AB9 Base. In doing so, I have encountered a problem with the SDK, which I have since been able to isolate and reproduce.

The effect I’m aiming for is very simple: when a certain angle of attack is reached, a constant force on the pitch/Y-axis should push the control stick forwards briefly, or later on continuously. This is intended to replicate the real stick pusher of the MiG-29’s angle-of-attack limiter.

Hardware/Software:

• MOZA AB9 Base
• Windows 11 x64
• Latest MOZA Device Service / MOZA Cockpit
• MOZA Flight SDK x64
• Visual Studio / C++ x64
• ApiDebugTool v1.0.0.4

Communication between my C++ programme, the MOZA SDK, Device Service and the AB9 is working. The Base is recognised correctly and an FFB effect can be successfully registered using MainCtrl_FfbNewEffect (Command 603, Type 1 = Constant Force). The returned effect index is 1.

The actual problem concerns MainCtrl_FfbConstant (Command 604) and, in particular, the enableAxis field.

The effect works perfectly with the supplied MOZA ApiDebugTool. I have tested the following values:

• effectBlockIndex = 1
• magnitude = 600
• duration = 200
• enableAxis = 0 / 1 / 1
  • X = 0
  • Y = 1
  • DirectionEnable = 1
• directionX = 0
• directionY = 0

I then start effect 1 via MainCtrl_FfbEffectOperation (Command 609, Operation Start).

Result in the ApiDebugTool:
The AB9 moves the control stick forward by approximately 0.5 cm in a reproducible manner. The desired constant-force effect on the pitch axis therefore works in principle.

However, when I send the same parameters via the C++ Flight SDK, no movement occurs.

The request generated by the SDK looks like this:

moza::as::wheelbase_type::MainCtrlFfbConstantRequest constant{};

constant.effectBlockIndex = 1;

constant.magnitude = 600;

constant.duration = 200;

constant.enableAxis.field0 = 0; // X

constant.enableAxis.field1 = 1; // Y

constant.enableAxis.field2 = 1; // DirectionEnable

constant.directionX = 0;

constant.directionY = 0;

base.cmdMainCtrlFfbConstant().send(constant);

In the generated header file `device_as.generated.hpp`, `enableAxis` is defined as a structure for Command 604:

struct MainCtrlFfbConstantRequestEnableAxis

{

    int field0{};

    int field1{};

    int field2{};

};

However, it is precisely when this field is serialised that SDK warnings regarding BitSet / structFields / type mismatch appear.

We have also attempted to send Command 604 via RawCommand. Here too, the SDK appears to convert the `enableAxis` argument based on the command metadata. For example, a test using a hypothetical bitmask value of 6 did not produce the desired result, and the payload actually generated did not match the expected value.

It is also striking that other FFB requests, such as Ramp, Periodic and Condition, use `enableAxis` as a simple `std::uint8_t` in the generated header, whilst Constant Force (604) uses this specific structure with three fields.

Hence my questions:

1. How must `enableAxis` for `MainCtrl_FfbConstant` (Command 604) be correctly set in the C++ Flight SDK so that exactly `X=0`, `Y=1` and `DirectionEnable=1` are transmitted?
2. Is the definition

MainCtrlFfbConstantRequestEnableAxis

{

    int field0;

    int field1;

    int field2;

} ;

in the currently generated `device_as.generated.hpp` correct?

1. Is there a specific bit-set encoding or a specific SDK data type for this field that we need to use?
2. Could you please provide us with a complete C++ example for Command 604 on the AB9 that generates a constant force exclusively on the Y/pitch axis?

The crucial point is:

The same values work in the MOZA ApiDebugTool, but not via the public C++ SDK call.

This allows us to rule out a fault with the AB9 itself or with the basic FFB function. We simply need the correct C++ control or serialisation of `enableAxis` for Command 604.

For a reliable test, the following is sufficient for now:

magnitude 600, duration 200 ms, Y-axis.

Increasing the force is not necessary for troubleshooting.

If it would be helpful, I can also provide you with the complete C++ test code, SDK debug output and device service logs.

Thank you very much for your support.

Kind regards,
Matthias Rausch