Time Acceleration and Heading Indicator Issues

[greggerm 312]

Good day, experts. Time for a good problem.I just performed a reinstall of FSX (with Acceleration) and I am running into an odd issue that I don't recall seeing in the past. Tested with the default Cessna...I can fly at 1x, 2x, and 4x speeds without issue. When I begin flying at 8x speed, the directional gyro ends up getting out-of-phase with the actual heading and zips along at double speed. When turning through a full 360 degrees, the gyro card will end up doing TWO full rotations. If I fly at 16x, the directional gyro ends up moving at 4x speed, and the gyro card makes a full 360 rotation when the airplane only makes a 90 degree turn. In all scenarios, my SHIFT-Z and GPS headings remain correct. Only the gyro gets messy. Gyro drift is disabled in FSX settings. Can anyone shed some light on this scenario? I'm not often confused by MSFS, but this one vexes me! -Greg

[greggerm 312]

Update:I restored my computer back to a "clean" build without any games or FSX installed at all, and started over. I installed FSX, did little to no configuring other than to get my video driver up to snuff, and tried it out. The directional gyro stayed rock-solid at any time compression setting. (pure, stock FSX)I took an Acronis backup for posterity, and installed the Acceleration add-on. Again, no additional configuring, just ran it up, jumped in the cockpit of the C172, and tried it out. Boooo, the directional gyro was only stable at 1x, 2x, and 4x - above that, it reverted to the behavior I saw earlier... at 8x time compression, the gyro zips around at twice its normal rate... at 16x, it flies at four times its normal turn rate. I took another Acronis backup for posterity.I restored the earlier, stock FSX backup and tried it once more - again, it worked fine there. The only difference between the two? Stock FSX seems to work, Acceleration seems to have an issue.Can someone else validate or test this theory? I don't recall having this issue in the past at all, but even after a fully clean installation, it's showing up! I'm stumped, and hoping to get some guidance.Thanks,-Greg

[greggerm 312]

If it's of any assistance, I am running Windows 7 Pro x64 on an i7-860 with 8GB RAM. I can't imagine that any non-MSFS applications would cause such a problem, but I'm all ears. I'd appreciate it if someone could test out their FSX-Acceleration install by loading up the default Cessna C172, flying north, then flipping on time compression up to 8x or 16x and watching their directional gyro as you enter a turn. I'd love to know if it remained on the right heading as compared to the GPS and/or SHIFT-Z heading indicator. As I mentioned, I never noticed this issue before, but that's not to say I haven't had it all along durig my previous installs. Your input is appreciated,-Greg

[WarpD 824]

I fail to see the importance of this 'test'.There is no validity to running the sim at that speed and then expecting everything to be 100% accurate.

[greggerm 312]

Huh. I didn't think I'd have to defend the legitimacy of a request for assitance, but... (shrug) Testing to see if this behavior is consistant across various installations will help me see if this is an inherent issue to FSX + Acceleration, or if it may be something that is affected by other variables. As I mentioned, I don't recall seeing this issue on my previous installations of FSX & Acceleration, so I am curious to find out if I introduced something along the way to cause it, or if it's "just the way it is" and I never noticed it before. I'm well aware of the effects of time compression against things like autopilots, but the directional gyro anomoly I observed isn't a "flutter" like the A/P trying to hold an altitude under the stress of time compression. It is a repeatable and fixed multiplication in the rotation rate of the gyro card, and it struck me as something a more systemic in nature than the computational inaccuracies introduced by the increased time compression seen in other scenarios. If there's nothing that can be done to fix it, so be it. That's fine. But if others do NOT experience this anomoly, I'd like to know so I can attempt to clean up my install somehow. If anyone can spare the couple of minutes to try it out, I'd still appreciate it.-Greg

[WarpD 824]

Let me rephrase... I don't see where this actually affects anything as the directional gyro isn't used by an autopilot and you certainly can't fly by hand at 16x+. So, how is this impacting you at a level where it's honestly important?If you are seeing it on a clean install with no non-Microsoft addons... then I would hazard that it's a flaw in FSX itself.

[greggerm 312]

Well, I've done some experimentation and reviewing of this issue and have some notes from some basic but enlightening tests:

• This does not affect all airplanes. Certain gauges are affected, but not all. Obviously the default C172 is affected, but the Beech King Air is not. Interestingly, the Flight1 Pilatus PC-12 is very much affected. Most other default and addon airplanes do not exhibit the issue.
• The "D" key to resync the directional gyro is effective in instantly clearing up the issue, although any turning while in compressed time will immediately cause the problem to resurface.
• Unlike the default C172, the Flight1 PC-12 has a problem with compressed flight and GPS-slaved autopilot flight. I'm not sure how and why, but when the time compression errors throw the gyro out of whack, the GPS course begins to deviate. It is almost as if the PC-12's systems are not flying via the GPS directly, but rather it uses the course designated by the GPS and it uses the directional gyro indication as the reference on which to turn. It's reading the heading off the GPS, but using the heading on the gyro to align to. When there are turns or mild course corrections while under GPS guided flight in compressed time, these corrections are amplified greatly on the directional gyro, leading to a continuously deviating flight. Liberal use of the "D" key to resync the gyro helps keep things on track. Other airplanes seem unaffected - only the PC-12 seems to have this issue.

I am certainly getting along OK with the issue. I don't know if anyone else has experienced these problems, as nobody seemed to give it a test, but that's OK. It is a pretty specific set of conditions which brings out the problem. That's it from here... -Greg

[Guest bratonabike]

Ill give it a test this weekend and get back to you

[PhilTaylor 45]

Ill give it a test this weekend and get back to you

If I recall correctly, anything past 4x doesnt get full simulation because the extra large step size starts causing numerical instability in the equations.So while you can run timesteps higher than 4x, dont expect the simulation to be accurate if you do. I see that feature as mainly good for long, international tubeliner runs in VAs across the empty oceans. Not as something one would do often, and not as something one would expect all elements of the sim to work during.

[jjjallen 37]

If I recall correctly, anything past 4x doesnt get full simulation because the extra large step size starts causing numerical instability in the equations.So while you can run timesteps higher than 4x, dont expect the simulation to be accurate if you do. I see that feature as mainly good for long, international tubeliner runs in VAs across the empty oceans. Not as something one would do often, and not as something one would expect all elements of the sim to work during.

I like to fly to EHAM on the red-eye in realtime. The best part is sunrise over Ireland.jja

[greggerm 312]

If I recall correctly, anything past 4x doesnt get full simulation because the extra large step size starts causing numerical instability in the equations.So while you can run timesteps higher than 4x, dont expect the simulation to be accurate if you do.

Indeed - it is interesting to me that the majority of airplanes I've tested do not experience this issue, so I don't think this is a global FSX situation. Perhaps the gauge files or some sort of offset the effected gauges look at are the root causes... and the calculation instability makes perfect sense as a potential underlying cause. You win some, you lose some I guess. Please understand that I certainly do not have any expectations of having this fixed or otherwise tweaked. The real intent of this thread for me was just to see if I was the only person experiencing it and under what conditions. I'm actually a bit happy that my research has shown it is inherent to FSX and not brought about by an addon or something.Phil - nice to see your name out here once again! I haven't noticed many posts in your name lately (*might just not be paying attention), but it's a pleasure to have you still participating after all you've done for the community! With that, I'll cancel my tracking subscription to this thread. I think we're all answered here! -Greg