F-16 Waypoints off by 20nm after 1 hour flight?

[MeanJim]

This might need to be posted to bugs, but I don't want to jump the gun just yet, plus I don't have a track file because DCS crashed and it was corrupted.  I do have the Tacview replay, but I doubt that would be of any help.  I noticed the patch notes for the recent update says: "Fixed: INS drift is not GPS corrected and INS drift adds up to more than 1000 ft over time."

I have been playing on the TTI Syria server lately, and have had no issues, until now.  Unless I get shot down, I land, re-arm, and refuel, and go back up.  I've never needed to do an INS re-alignment, and I've only noticed a small amount of drift over several hours (2-4) in a session.  Last night though, after 1 hour my waypoints were 20nm off.  I don't know if this is an INS issue or something else, because it was very odd.  The location of the waypoints in the HUD, HMD, EHSI, and the cake symbol on the radar all showed the waypoints in the wrong place.  The HSD page showed waypoints in the correct place, and the auto pilot navigated to the right place in steering hold.

I didn't notice anything was wrong until I was RTB.  I had a quite a way to go, so I flipped the auto pilot to altitude and steering hold while I cycled through other players to watch them while my jet cruised home.  I popped back into my cockpit occasionally, and 50nm out I switched to attitude and steering hold and started a descent.  When it showed ~25nm out I realized something didn't look right.  The waypoint symbol was outside of the HUD to my right.  I looked and the waypoint symbol in the HMD was in the middle of nowhere, not an airport.  I looked around, and the city was to my left, so I knew I should be near the airport.  I looked at the HSD and it showed I was passing over my home waypoint.  I banked and I looked down and sure enough I was right over the airfield.

I landed, and taxied to a parking area where players don't spawn to re-arm and refuel.  After I re-armed, I performed a full INS re-alignment.  I verified that the INS page showed the correct LAT/LNG and ALT before aligning.

With 8 minutes to kill, I cycled through other players until the alignment was done, but it still showed my waypoint 20nm north of where I was, when it should have been right in front of me.  I know I could have re-slotted and been on my way, but I was determined to fix this.  I tried doing a complete shut down and restart of the jet, and another full INS alignment, but it still had my waypoint 20nm north of me.

After that, I had no choice but to re-slot and start over with a fresh jet.  My second  mission was a little bit longer because while I was messing with all of the above, other players had cleared the area I was working on of SAM threats and were mopping up the remaining ground units.  The second flight was about 1.5 hours, and upon returning home, my waypoint was off again.  This time it was 10nm northwest, not as bad as before, but it shouldn't be off by that much in such a short time.

When I disconnected, DCS crashed, which I don't think has ever happened either.  I did submit the crash when the crash reporter popped up.  The track file is probably corrupted because of the crash.  I checked the tracks directory, and it has two files of the same name, but one has the .trk extension, which is only 11MB, and a .trk.writing that is 10MB, both are way to small for as long as I flew.  I did try anyway though.  The .trk fuke gives an error, and renaming the .trk.writing to just .trk causes DCS to sit at an infinite loading screen until I kill it.

This screen shot shows the LAT/LNG on the INS DED page, which matches what I get from the F10 map.  I ALT-clicked my aircraft on the F10 map to get the popup that stays when you go back to the cockpit, but I wasn't thinking and closed it when I took the screen shot.

I took this screen shot to show that the position of the waypoint should be in front of me, I don't know why the map is in MGRS, but you can convert it to see that it matches what is shown above.  You can also see that the waypoint is correct on the HSD.

This screen shot is me looking over my right shoulder where it's saying the waypoint is.

[Tholozor]

Looking at the first screenshot, you forgot to Cursor Zero (denoted by the CZ option on the HSD).

[Mr_sukebe]

Thoughts:

- in an hour you might well have travelled 500 miles.  20 miles variance is a 4% error for an Ins system.  I’ve no idea how accurate real ones are, but that’s just a thought

- as I understand it, INS is accuracy tends to worse with a lot of manoeuvring, that might not have helped

- I can’t remember if the INS in the F16 accounts for wind. Even if it does, it won’t be 100% accurate

I might well be wrong, but I wouldn’t be surprised if that’s common for INS systems.  
One the bright side, the F16, does have a way to update accuracy on the fly using known preplanned map locations.

[Rongor]

5 hours ago, MeanJim said:

This screen shot shows the LAT/LNG on the INS DED page, which matches what I get from the F10 map.  I ALT-clicked my aircraft on the F10 map to get the popup that stays when you go back to the cockpit, but I wasn't thinking and closed it when I took the screen shot.

I took this screen shot to show that the position of the waypoint should be in front of me, I don't know why the map is in MGRS, but you can convert it to see that it matches what is shown above.  You can also see that the waypoint is correct on the HSD.

This screen shot is me looking over my right shoulder where it's saying the waypoint is.

 

Too bad your F10 map screenshot doesn't show LL, you chose to display MGRS there...

Unless TTI is running a date before 28 March 1994 (top right corner of F10 screenshot shows they do not) and/or you didn't turn the INS mode knob to NAV after alignment, there is no reason to assume you would see any noticeable drift at all, as the INS is a blended solution constantly combining INS and GPS.
Read more on this in the linked article below.

 

Yet as your INS DED page confirms, the INS does know where you are, so the offset of steerpoints will most likely have a different reason.
@Tholozor's suggestion is pointing into the obvious direction. Always look out for CZ in your MFDs and press it when you require steerpoint accuracy again after playing around with some SOI (read more on this on page 303 in ED's manual)
Especially use of HTS can easily cause offsets in the 20 NM range.

Edited Saturday at 08:55 AM by Rongor

[MeanJim]

4 hours ago, Tholozor said:

Looking at the first screenshot, you forgot to Cursor Zero (denoted by the CZ option on the HSD).

Hmm, that appears to be what it was.  I don't know how I missed that, but it brings up more questions.  Why did shutting down and restarting the jet not clear the cursor slew?  That screen shot was taken before restarting, so I don't know if CZ was still even an option after that.  That should be easy to test though.

I had forgotten there was even a CZ button on the HSD because I've made it a habit to press TMS down a couple of times before I switch from A-G mode back to NAV mode to undesignate any SPI and reset the cursor.  I was flying long range SEAD missions that night, and only had the HTS, two HARMs, two fuel tanks, and some AAMRAMs.  Reading back over the HAD/HTS/HARM pages in the manual, it sounds like it's not supposed to reset the cursor, but I haven't had this happen in ages.

[Rongor]

46 minutes ago, MeanJim said:

I was flying long range SEAD missions that night, and only had the HTS, two HARMs, two fuel tanks, and some AAMRAMs.  Reading back over the HAD/HTS/HARM pages in the manual, it sounds like it's not supposed to reset the cursor, but I haven't had this happen in ages.

page 407

[MeanJim]

That's the page I was talking about.  It says that rejecting a target on the HAD doesn't clear the sight point.  I've been pressing TMS aft before switching back to NAV mode and it has been resetting the cursor back to the steerpoint, and it shouldn't, so maybe it was a bug that they fixed.  From now on I'll just remember that doesn't work on the HAD.  The email that came a day after the patch says "a long list of HAD and CMDS improvements", but I only see two things listed in the change log related to the HAD.