OnCourse Software

George,

I've made some changes to our PF3_Validate program, which will now check to confirm each of the DLL's and ActiveX controls installed by PF3 are actually registered correctly.

Please unzip the attached into your main PF3 folder and either run it direct of via the PF3 Options #2 menu

Good news and bad news. The program runs indicating there are no problems, that is also the bad news. I've included two files:
1.Validate-StartupLog.zip is the validate log and startup log as a result of installing the program and running it off of the option #2 menu.
2. Validate After Boot.zip is just the validate log from a standalone run of the validate program after a boot of the system. I did not boot before running the program the first time.

First programming was Fortran IV on the IBM in college and first PC was Level 1 Basic on the TRS-80 Model 1 both circa 1979. I worked at Radio Shack part time. Heady days indeed!

Ready for that Macallans yet?
Cheers,
George

Well I'm fast running our of ideas!

Please remind me again what AV and other security programs you are running. I'm sure you've already mentioned it but (a) I can't remember and (b) and too lazy to scroll back thought this 7 page thread!

If the DLL's are registered correctly, which they appear to be, then the problem is when PF3 attempts to create an object... that's the poit the 91 error pops up. So is it possible you have some system preventing this to happen due to privileges, etc? Clutching at straws I know.

Perhaps the next thing to try would be (1) disconnect your PC from the Internet (2) terminate all running programs such as your Av and any other security type apps (3) see if the PF3 error continues

The only security program I run is Windows Defender and I have the PF3 folder excluded from the AV scans. I removed both Bitdefender and Acronis Backup as I saw what appeared to be problems from them early on regarding ransomware defense. Both programs raised hell, or appeared to, when installing FSX:SE after installing Windows 10. A bummer as I've spent a fair coin on them and liked their features.

I unplugged the Internet, turned off Windows Defender, both via the UI and went into CCleaner and turned off the program SecurityHealth (MSASCuiL.exe). No difference.

Here's the thing. Why do some flight plans work and some don't? I would (mistakenly) think that either they all work or all fail. But you have loaded up my flight plans and they work just fine. Why only in certain instances is there a problem creating an object? I know, the $64K question.

Your continuing effort is appreciated, thanks again,
George

7X7VirtualPilot wrote: *

Wed Aug 30, 2017 10:44 pm

Here's the thing. Why do some flight plans work and some don't?

No real ideas, just continuing to guess:
Are you doing exactly the same things after loading them? (i.e. setting SID/STAR marks, adjusting cruise altitude, ..., or nothing of these things)
Do the affected flight plans work if you don't open the altitude adjustment screen? (you can disable automatically opening it after loading the flight plan, or maybe that is even the default)

Exactly the same things. If I don't attempt to Adjust Altitudes, SIDs, STARs and Holds, there is no problem. Nothing different there. My guess is that you're setting up a table(s) when you adjust the altitudes, etc and there is something about that process that is abending (is abending a mainframe term? ). But Dave has pretty much already said that. But why only certain flight plans? Is there something about the number of waypoints? I understand it works for you Dave, but still the size of the table may be an issue with my 'release' of the opsys. This 91 issue is the critical factor keeping me from easily using PF3. Dave, if at some time you want to jump on Skype and discuss (I can aim my camera at the screen so you can direct the action), am up for that. Might be better than this marathon thread. Just a thought.

Cheers,
George

7X7VirtualPilot wrote: *

Thu Aug 31, 2017 9:37 am

Exactly the same things. If I don't attempt to Adjust Altitudes, SIDs, STARs and Holds, there is no problem. Nothing different there. My guess is that you're setting up a table(s) when you adjust the altitudes, etc and there is something about that process that is abending (is abending a mainframe term? ). But Dave has pretty much already said that. But why only certain flight plans? Is there something about the number of waypoints? I understand it works for you Dave, but still the size of the table may be an issue with my 'release' of the opsys. This 91 issue is the critical factor keeping me from easily using PF3. Dave, if at some time you want to jump on Skype and discuss (I can aim my camera at the screen so you can direct the action), am up for that. Might be better than this marathon thread. Just a thought.

Cheers,
George

No, it has nothing to do with setting up tables, etc. When you hot that 'Quick Edit' button the code simple creates an object t create a tool tip and that's what kicks the 91 error 'object variable not set', which of course it is. As mentioned earlier, if the DLL wasn't registered correctly you would get the same error.

I don't use Skype but I'm happy to take a remote look at your system using TeamViewer?

7X7VirtualPilot wrote: *

Tue Aug 29, 2017 4:32 am

Dave,
Didn't have time to work on it today. Hopefully tomorrow.

Question for David: Is yours a pure Windows 10 PC? Did it ever run anything other than Windows 10 Creator's update? This is why I intensely dislike M$, this operating system as a service idea. I'm getting too old and tired to fight them, 32 years is too much in my book. If only FSX was Linux based. Sigh.

Good to know Dave there are still some dinosaurs still roaming the earth. I dabbled in Pascal, Modula-2, Prolog, SQL and the craziest of all, LISP. Haven't seen the advantage in the modern languages like C#, C++, PHP, Python, Java etc, etc. How many ways can you build a language? I still like ol' K&R C. Simple and powerful. I should take a look at VB. But I'm retired, so there's that, lol.

Thanks Dave and David, tomorrow will try, try again.
George

Hi George,

My PC originally ran the base Windows 10, I did both the anniversary upgrade and the Creator's update.
The PC was built 2 years ago and always ran some sort of Windows 10.

I see about setting up an object, but is that different for different flight plans? And on a "bad" flight plan, it is one of three actions that cause the problem.
1. Button to Set END of SID/DP
2. Button to Set START of STAR
3. Check the box to Activate quick-entry mode
Any one of those actions will cause a "bad" flight plan to get an immediate '91'. So far they will NEVER create a '91' on a 'good' flight plan. Absolutely consistent. My low level way of expressing setting up of a table should have been phrased as creating an object. Those activities are done at a higher level now, but underneath that is what you're doing. When you create such an object, do you pass any variables to the routine that creates the object? Like number of rows (waypoints or whatever) that are expected to be processed? I'm trying to understand if there is something in memory management that is having a problem with some size that is present on some flight plans but not others.

I'm up for whatever puts this to rest. I've never used TeamViewer but if you hum a few bars...(twist of an old joke).

Dave, could you create a test version that disables the tooltips for those fields/buttons? This way you could see if this is the only problem, and a permanent workaround could even be to allow disabling tooltips.