Posts by aenbacka

    Isn't Autodesk Maya like a $5000/seat commercial software like 3DS Max? Man, I would just stick to using Blender unless you are thinking to become a payware aircraft model developer but I doubt you would make your investment back in sales. I too would be interested in hearing what @Jet-Pack has to say about this. I know at one time long ago both Laminar Research and IPACS used to have converter plugins for other 3D CAD programs but I think they eventually switched over to only supporting Blender. I am guessing because it is free and very full featured (actually more than you need probably for just flight simulator models), and in the case of X-Plane encourages hobbyists to try making their own airplanes for their favorite flight simulator. That was a very smart move on Laminar Research's part. That and having a GUI based aircraft file editor called Plane Maker is why X-Plane 11 has literally hundreds of freeware aircraft available for it including helicopters. In fact, from what I have read in the forums, if you just want to test out an idea for your own custom aircraft design, you don't even need a 3D CAD program with Plane Maker because it can create a very basic 3D aircraft shape which you can then fly in the X-Plane simulator.

    Yes, Maya is commercial. I am using Blender but I have people helping out that uses Maya/3ds max. The exporter should work with 3ds max and Blender, so if we can get the models into one of those (including animations and rigging) it should be able to export to tgi. I tried to export the Carbon Cub model using the Blender plugin (using Blender 3.3 lts) but I got lots or error messages and exporting failed.

    Please check if the copilot.tmc files have the same set of parameters for the engines. EngineCount should be 1, Piston should be true.

    Thanks a lot, that was it :) I did not think about that the copilot.tmc file could have any impact. With the suggested additions, Controls.Throttle1 works correctly now.

    Jet-Pack (IPACS) I finally got the throttle to respond by doing some debugging (manually setting the different inputs to a fixed value, to trace the entire chain). It seems the problem is in the input_lever definition, this code works correctly:

    Code
    // throttle
    <[input_lever][ThrottleInput][]
    <[string8][Message][Controls.Throttle]>
    <[string8][InputEnable][1.0]>
    <[tmvector2d][Range][ 0.0 1.0 ]>
    <[float64][Value][0.0]>
    >

    If I use Controls.Throttle1 (which the Aerofly FS2 to FS4 documentation suggests), the throttle does not work at all in this particular aircraft. What could be the reason for this?

    I did a test by setting engine InputThrottle to a fixed value 1.0, and that seem to give full throttle, so something with the throttle control input.

    Is it possible using the developer mode to see the different inputs and outputs in the tmd, as the infobar contains quite limited info for debugging?

    Replace <[string8][InputMixture][ActualMixture.Output]> with <[string8][InputMixture][1.0]> and check if that fixes the issue

    Tried with InputMixture 1.0, but still no luck. The engine2 definition is now as follows:

    And for the throttle I have:


    My first guess is that the engine input cutoff is on, which disables the engine. This could be some cold and dark hack from the original aircraft.

    What tmd element determines the cutoff? I used windiff to compare tmd files of pipercub and carboncub, but I do not spot any differences which would explain why the throttle does not increase / decrease. Rudder control works properly, and if starting in the air it shows 70-80% throttle.

    The easiest way to get something to work is by copying it from another aircraft. A lot of messages were added for Aerofly FS 4 which could not be assigned in Aerofly FS 2.

    We do have a 3D editor to visualize parts of the TMD. I can send you a copy, though the tool itself isn't documented. But I'm keeping it up to date because that's what we're working with as well.

    Yes I have now tried to do some comparisons with e.g., the DR400, but for the throttle I have not spotted any differences. It would be great if you could send me the tool, very much appreciated😀

    Working with the TMD file requires quite some effort, especially as the information on the Aerofly does not seem to be completely up to date, or missing (some sections do not exist at all, for instance part of the engine section).

    Also the porting instructions from Aerofly FS2 mentions that messages Controls.Starter etc should be renamed to Starter1.., however it is slightly unclear if this concerns all Controls messages (such as Controls.EngineStarter).

    Also is there any more detailed debug output so you could see the different Inputs and Outputs in realtime during development, as that would be extremely helpful?

    Went through and compared several tmds this weekend, but did not find any difference that would explain why the throttle input is working.

    The engine parameters between the piper cub and carbon cub seem similarly defined, and also the throttle_lever / input. If starting the simulation in the air, the throttle is active (around 70%) but not possible to increase or decrease the value. Touchdown seem now improved after the propellerbody adjustments.

    Before takeoff or landing flight testing, I would focus on ground testing - being able to accelerate and slow down on the tarmac, being able to turn left and right on the tarmac, being able to brake to a stop. Currently, the only way to know your engine is running and under positive control is being able to speed up and slow down in response to throttle position. If you had an RPM instrument or engine sound setup then would also help. At the same time I would suggest getting your tailwheel, ailerons, flaps, elevator, and rudder graphics tied into the tmd file and tested on the ground before flying (both for correct deflection limits and direction sense and response speed). Regardless of the real world aircraft free castering or not steering system, please always implement either nosewheel or tailwheel steering since a lot of FS 4 users are not going to have differential braking rudder pedals controls. I would get the engine sound connected and create the RPM instrument frame animation setup in you cockpit panel as engine RPM is not available in the flight information bar display. Throttle position does not really correlate well to engine RPM like 50% throttle is not going to give you 50% of the max RPM.

    Yes my focus now is on getting the throttle and rpm control to work, and continue the work with the model on the ground. The interior is in progress and hopefully we can get some instruments hooked up to the simulation soon. The tail wheel is controllable using rudder right now, and the animation seems to move nicely in reponse to rudder control.

    Also, from what I can understand from glancing athe aircraft:tmd:instruments wiki airspeed indicator discussion

    https://www.aerofly.com/dokuwiki/doku.…tmd:instruments

    "The output is differentiated to get the airspeed trend and the airspeed and the airspeed trend are then send to external devices or the flight info bar via sender objects."

    So if you include a sender for the throttle you should see your throttle position on the "I" key flight info bar, to see if you throttle command is changing.

    Code
                <[sender][SenderThrottlePosition][]
                   <[string8][Input][ThrottleInput.Output]>
                   <[string8][Message][Aircraft.Throttle]>
                >

    I do have that sender defined in the tmd, and from that it is visible that throttle do not change at all.

    It also may be that you have not defined all the tmd:engine2 parameters correctly so that maybe your engine is running but you cannot actually control it yet?

    I suggest experimenting with Kaniewski's Club Special engine2 settings to see which parameters must be defined to have control of the engine rpm.

    Yes the engine seems to be running, but I guess it could be some parameters that cause problems. Jet-Pack (IPACS) Do you have any ideas what could cause this? Now I have been able to touchdown multiple times without crashing, seemed at least some R0s related to propellerbody and joint where defined in such a way, which could cause ground contact. Slow progress, but at least some progress :) Seems to take time to get used to the tmd concepts.

    Also, I would ask Jet-Pack about what information can be on-screen display enabled during flight for developer debugging. There are some settings in the main.cfg that might allow you to specify what data you want to see via the Ctrl+F1 and Ctrl+F2 developer debug display commands. For instance, it would be very good to see engine rpm so you could tell if the engine was still running after the crash.

    Thanks I need to check the dev settings, if I have them enabled. It would be very useful if you could display the rigidbodies during simulation and debugging, so it would be easier to see if there were any contact with ground. Also, the throttle is strange, it seems to be defined similarly to other aircraft but it does not react at all to control inputs.

    Below is a link to a Youtube video where I am testing short final with the Carbon Cub; even though the set down seems not too rough the sim seems to reset. Is this also due to propeller hit ground (even though it seems to be no contact)? Also, of some reason the throttle does not respond on startup at all (always 0%), what could be the cause of this?

    Regarding the videos, they are very interesting to see and I am glad you have a Youtube channel. If you don't mind my asking, when you get to the point of creating your instrument graphics and animation in Blender, could you post a video showing how you make a steam gauge instrument in Blender, like maybe an example of doing either the RPM or VVI instrument and how you tie that into the aircraft tmd file?

    Yes I have been trying to add some videos from different sims there. Will see if some videos can be added concerning instrument development, once I get to that point. I try to get the model to a state in which is somewhat stable in the sim, and then start looking at the details.

    Should the following elements be correct for getting the throttle input to work correctly (if I have some controls mapped to the Throttle command)?