Most of the commonly reported issues and questions are answered in the Frequently Asked Questions (FAQ) option under the Support menu of this website.
*** PLEASE NOTE ***
Your forum account is not the same as the account used in the shop. They are completely separate accounts.
1st December: A new version of the GoFlight Interface Tool for MSFS is now available.
12th November: A new version of the GoFlight Interface Tool for X-Plane is now available.
6th June: A new version of Virtual Flight Sim Hardware is now available (huge update)
21st November: A new version of the GoFlight Interface Tool for FSX/FSXSE/P3D is now available.

GF-MCP Pro/GF-166/GF-46 work well with A2A C172 but repeatedly get stuck for 5sec with PMDG 737

I have P3D V4 with A2A c172 and PMDG 737NGX installed. I'm using latest version of GIT 3.3.0.1.

When using A2A C172, all GoFlight instruments are working correctly - dials and displays update immediately without any lockups. If on the other hand I select PMDG 737, the GoFlight instruments work but only for 2-3seconds at a time followed by about 5-10seconds pause where no inputs are received from the instruments. If I switch back to A2A C172, this bad behavior continues - I need to restart P3D and GIT for A2A C172 to not have this problem.

It feels like something in GIT gets stuck.

I wanted to check which variables and events and any scripts that are used with PMDG, but whenever PMDG aircraft are selected, Detective is greyed out as is the event menus for MCP Pro and EFIS.

Any suggestions will be most welcome.

Answers

  • Normally an uninstall and reinstall of the PMDG aircraft fixes issues people are having. I suspect all the recent PMDG updates and messing things up for some people.

    GIT does need the later versions of the PMDG aircraft to work.

    Detective is disabled for all PMDG aircraft as the SDK is built into GIT and Detective isn't needed.

    Best wishes
    Steve
  • I removed PMDG and also the pmdg.xml file my documents\goflight ... I reinstalled PMDG, updated everything using PMDG control center, restarted, activated MCP in GIT. At first I didn't have the freezing problem reported here - I could use the 166, 46 and MCP dials without any temporary freezing; however most of the displays in MCP showed 0 even though the dials were updating the instrument in the virtual cockpit.

    The second time I loaded PMDG 737, I received the message from GIT saying that it had loaded the latest PMDG information and that I should reload the aircraft. When I reloaded the aircraft, I was back to the same problem where moving the course dials would cause freezing for 5-10sec.

    How can I prevent GIT from loading the PMDG config and instead do it myself by knowing the various internal variables etc?

    Marc
  • The message you received would have been from PMDG. They added that warning as they know it breaks their SDK and thus add-ons which use it. I would suggest you go through the steps in the FAQ item called "My MCP PRO or MCP Advanced is not updating its display (just showing zeroes)" here: https://www.pollypotsoftware.org.uk/support/user-guide/user-guide-git/16-installation/70-faq-goflight-interface-tool.html

    You also need the latest version of the NGX (1.20) and the hotfix version of P3D.

    You cannot build your own config for the PMDG.

    Best wishes
    Steve
  • Hi Steve,

    Thanks for the info.

    I went ahead and check my versions:

    - P3D v4 4.0.28.21686 (this includes hotfix1)
    - NGX 737 1.20.8443

    I checked 737NGX_Options.ini and under [SDK], EnableDataBroadcast=1.

    I checked that in GFConfig that MCP Pro is configured for Compatible Add-On set.

    So it looks like everything is as required based on the above pointers.

    I made the following observation. I set EnableDataBroadcast=0 and started P3D and loaded PMDG- all GoFlight instruments work correctly without freezes with the only disadvantage that the digital displays of MCP Pro display nothing.

    Then I closed P3D, changed EnableDataBroadcast=1 and started P3D and loaded PMDG - I get the same problem where rotating the COURSE dial will freeze every few seconds for about 5-10sec. When this happens, all inputs from all other GoFlight instruments is blocked during that time. However during that time, I can update the dials on the virtual cockpit. The other thing I noticed is that if I turn the COURSE dial on the virtual cockpit, the MCP Pro display updates and doesn't get stuck. So then I decided to increase the write and cmd delay and that solved the problem.

    I wonder why I don't get this when I use the MCP and other instruments with other aircraft? Perhaps because PMDG is updating the data faster and this causes us to hit timing problems between reading and writing data between the instruments?

    Marc
  • You should not have needed to do that. Did anything show in the GIT Log tab when the slowdown occurred?

    Thx
    Steve
  • Hi Steve,

    So it turns out that I only solved part of the problem. By setting CMD and WRITE delay to 0 for MCP, now the COURSE changes do not cause 5-10sec freezes for MCP operations. However the moment I press any button or any dial on the GF-166 or GF-46, then everything freezes for 5-10sec, including switches on SECM. If I move switches on SECM, I don't get freezes.

    It appears that any operation that is going to cause updates on displays on GoFlight instruments will cause all operations to freeze for 5-10sec. For example, I configured a dial on GF-T8-1 to update the NAV1 standby frequency that is shown on a GF-166. If I move that dial, everything freezes (including MCP) for 5-10sec. If I disable the GF-166, then the dial works correctly without freezes.

    I don't get any such behavior with A2A C172, including MCP operations.

    It looks like there is something in the behavior of the interaction between GIT and PMDG regarding the update of certain operations (seems to be things related to COURSE and COM/NAV) that introduces a delay when the result of those changes needs to be read back and displayed on an instrument, like something in the software is waiting for feedback and doesn't continue until it gets the feedback or a timeout is reached.

    Any idea how I can debug this? I am a software/hardware engineer so perhaps I can help in more depth.

    Marc
  • edited September 2017
    Can you copy and paste the log after the slow down starts. You can add /log to the filename properties of the icon if copy and paste doesn't work.

    I've got a feeling that something is borked in the 8443 release of the NGX as several people are reporting problems now with this version.

    Thx
    Steve
  • Here is the log message that occurs just before and after I turn one of the GF dials that increments NAV1 standby frequency:

    18:40:09 SimConnect: Event Received: SIM, 1
    18:40:09 SimConnect: Event Received: PAUSE, 0
    18:40:10 Aircraft Connection Started
    18:40:10 Simconnect Version: 3.4.0.0
    18:40:10 Aircraft FullName: PMDG 737-800NGX WL
    18:40:10 Synchronising Switches & Selectors .....
    18:40:13 Synchronising Switches & Selectors Complete
    18:40:57 SimConnect: SimObject Data Handler: Nullable object must have a value.
    18:40:57 SimConnect Message Processing Finished
    18:40:58 SimConnect Connection Starting.....
    18:40:59 SimConnect Message Processing Started
    18:40:59 SimConnect Connection Started
    18:40:59 Aircraft Connection Starting.....
    18:40:59 SimConnect: Event Received: Aircraft Loaded/Flight Loaded
    18:40:59 SimConnect: Event Received: SIM, 1
    18:40:59 SimConnect: Event Received: PAUSE, 0
    18:41:00 Aircraft Connection Started
    18:41:00 Simconnect Version: 3.4.0.0
    18:41:00 Aircraft FullName: PMDG 737-800NGX WL
    18:41:00 Synchronising Switches & Selectors .....
    18:41:03 Synchronising Switches & Selectors Complete

    As you can see, there is a NULL data object in there and this causes GIT to reset the simconnect connection, hence the "freezing".

    I don't have this with other non-PMDG aircraft.

    If you need further debugging, let me know.

    Marc
  • Thanks for this. I suspect a problem with Simconnect on your system, but I will track the error message down to confirm exactly what is happening.

    Best wishes
    Steve
  • edited September 2017
    Hi Marc,

    The error is indicating that the data message received from Simconnect is null and it never should be. Since other aircraft work OK, I think we can safely eliminate any Simconnect issues. It does look like a PMDG aircraft installation issue.

    I would uninstall the NGX, install using the latest installer from PMDG as Administrator and DO NOT APPLY UPDATES. Try GIT a few times and see if the issue remains. If OK, apply an update and try again etc. Let me know how you get on.

    Best wishes
    Steve

    PS: Remember the first time GIT runs, it updates the PMDG settings file to enable data transmission, so GIT and P3D need restarting after the first run.
  • Hi Steve,

    I tried exactly as you recommended. I actually removed GoFlight, GIT and PMDG, including removing files under My Documents, Users\...\AppDat and Program Files. I reinstalled and ensured that PMDG didn't update NGX. I enabled MCPPro first and there was no problem. Then I enabled one of my 166, and the problem was back with the same simconnect message.

    Anything else I can do to debug?

    Marc
  • Hi Marc,

    I think I have found the problem whilst working on the code at the weekend. I can get an Beta version to you to try or you can wait a few days for the actual release to come out.

    Best wishes
    Steve
  • Hi Steve,

    Really appreciate the effort. I can wait until your next release.

    By the way, I heard that P3D v4 has an update coming, hopefully this will not break things for you.

    Marc
  • Answer ✓
    Hi Marc,

    New release is out. Hopefully this will work for you.

    Best wishes
    Steve
  • Hi Steve,

    I finally got around to checking GIT 3.4.1.0. It fixes the freezes that I have reported in this thread. Unfortunately I now have problems with other aircraft - will start a new thread for that.

    Marc
Sign In or Register to comment.