Jump to content

INS only CCIP in the Hornet


Recommended Posts

yeah this seems broken(tm)

 

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

On 7/27/2021 at 7:59 PM, IvanK said:

Ok just had a look at INS drift over 30 mins POS INS only. after 30 mins the Drift was as shown in attached screen shot. So not to bad. However CCIP release resulted in bombs impact as shown by the X. I then attempted another CCIP release on a heading 90 degrees from the first ... same issue. Then landed GS showed 0 when parked. So something is very wrong imo. The Drift exhibited(though I consider ok)  should not have resulted in the CCIP bomb errors seen.

 

Progtgt.jpg

 

 

Impact.jpg

Doing an IFA is a not in real world a real option just to improve things ... it wont your original alignment will always be way better than any IFA. IFA is a last chance type thing to try and get something back after after say an INS crash etc.  A vel update using PVU might be a thought .... though again should not be required.... and it doesn't look like its implemented.

 

Either way the CCIP release impact errors seen after just 30 mins of flight are completely wrong. I tried an Auto release but need to try again after refing the designation to see where the bombs go.

 

Anyway its a major issue imo ... good pick up Svend.

 

CCIP is wrong in all modes all the time. I reported this bug here:

 

Take a TPOD and make sure no target designated so you can see where the TPOD thinks the bombs will hit vs where the CCIP cross is located. The TPOD shows you the correct aimpoint.


Conclusion: CCIP cross is mis-placed in all modes.

 

EDIT:

Looking back on this thread, it looks like we have two bugs, one on top of the other (possibly 3)

1. The CCIP cross is naturally displaced from the actual aimpoint, usually to the left. This occurs in all modes and is a confirmed bug. @IvanK this bug could explain the miss in your first test. Run it again with a TPOD like shown in the track in the bug report above and you can confirm.

2. The INS system is picking up some sort of lateral displacement it shouldn't in the absence of GPS.

3. (possible this is the same as 1 manifested differently) AUTO bomb fall line will "jump" around the TD box depending on airspeed and altitude, with certain combinations being just wrong and leading to misses. This happens in all modes.

 

It is possible 1 and 3 are exacerbated by 2 and vice versa. I've said it once and I'll say it again the iron bombing needs to be the first thing ironed out in these jets when they roll out and they need to stay tight, as they expose many of the other issues around the programming of how the jet works in sim. Things AA modes and the like may not exhibit because of the way they must be coded.

 

Also, there is a bug with Mk83's I reported that is over a year old that was never fixed. If you are not at 7000k and 370kts, you will miss long and short unless you adjust your altitude and airspeed accordingly. Lower = faster, higher = slower. Too low/fast you will miss long too high/slow you will miss short. So don't use Mk83's in your testing if in-line impact measuring is important to the test, they'll confound your results.


Edited by LastRifleRound
Link to comment
Share on other sites

Overthinking this I 🙂 I wondered if this was some error in Schuler loop modelling (if it is in DCS and if the DCS world is not flat). So repeated the experiment to be back on tgt after 84 mins flight time (schuler cycle time). Well All I got was huge errors in the same direction as before check out the CCIP (reflected cue) and DIL line now 🙂

 

Schuler.jpg

 

 


Edited by IvanK
  • Like 1
Link to comment
Share on other sites

Posted (edited)

I'm thinking about doing some tests to night where I will use the A/G radar to designate a point on the ground and store it as a mark point. 

Then fly a bit more, make a new designation on the same point and store as a new mark point until I have a bunch.

Then plot the mark points on the map to see what the trend is.

Dunno if that would make any sense. We can clearly see that something is off.

 

EDIT: come to think of it, it might not make any sense since drift is affected by maneuvers.

 

EDIT 2: Good news! after fiddling around with the A/G radar for a while i found out that if you actively track a ground object/target with A/G radar you can land WALLEYE, GPS guided munitions ( in 1981, how?) and even dumb bombs if you track the CCRP fall line on run in and forget about the diamond.

So A/G radar doesn't seem to be affected by INS drift even though it should output the wrong coordinates to the receiver.

 

Bad news is that you won't actually know which target the radar is tracking if area is tightly packed unless you have a TDP, maverick or walleye.

 


Edited by Svend_Dellepude

[sIGPIC][/sIGPIC]

Win10 64, Asus Maximus VIII Formula, i5 6600K, Geforce 980 GTX Ti, 32 GB Ram, Samsung EVO SSD.

Link to comment
Share on other sites

11 hours ago, IvanK said:

Overthinking this I 🙂 I wondered if this was some error in Schuler loop modelling (if it is in DCS and if the DCS world is not flat). So repeated the experiment to be back on tgt after 84 mins flight time (schuler cycle time). Well All I got was huge errors in the same direction as before check out the CCIP (reflected cue) and DIL line now 🙂

 

Schuler.jpg

 

 

 

Do you still have the track of your test?

If yes could you post it in the bug section so ed could fix it?

Link to comment
Share on other sites

13 hours ago, LastRifleRound said:

CCIP is wrong in all modes all the time. I reported this bug here:

 

Take a TPOD and make sure no target designated so you can see where the TPOD thinks the bombs will hit vs where the CCIP cross is located. The TPOD shows you the correct aimpoint.


Conclusion: CCIP cross is mis-placed in all modes.

 

EDIT:

Looking back on this thread, it looks like we have two bugs, one on top of the other (possibly 3)

1. The CCIP cross is naturally displaced from the actual aimpoint, usually to the left. This occurs in all modes and is a confirmed bug. @IvanK this bug could explain the miss in your first test. Run it again with a TPOD like shown in the track in the bug report above and you can confirm.

2. The INS system is picking up some sort of lateral displacement it shouldn't in the absence of GPS.

3. (possible this is the same as 1 manifested differently) AUTO bomb fall line will "jump" around the TD box depending on airspeed and altitude, with certain combinations being just wrong and leading to misses. This happens in all modes.

 

It is possible 1 and 3 are exacerbated by 2 and vice versa. I've said it once and I'll say it again the iron bombing needs to be the first thing ironed out in these jets when they roll out and they need to stay tight, as they expose many of the other issues around the programming of how the jet works in sim. Things AA modes and the like may not exhibit because of the way they must be coded.

 

Also, there is a bug with Mk83's I reported that is over a year old that was never fixed. If you are not at 7000k and 370kts, you will miss long and short unless you adjust your altitude and airspeed accordingly. Lower = faster, higher = slower. Too low/fast you will miss long too high/slow you will miss short. So don't use Mk83's in your testing if in-line impact measuring is important to the test, they'll confound your results.

 

 

TBH, I'm not quite sure what changed since then but currently I'm able to whack the target in AUTO or CCIP regardless of target altitude, as long as I'm in a dive (I only tried between 15 and 20 deg. dive at M .8 )  Level deliveries in auto are off, mostly short and left (3 out of 4)  I used 1 MK83 on station 5 (center) with ATFLIR on a cheek.

I'm gonna do some more tests today with HMD target acquisition and FLIR.  So far I tried to set up a target at 10k feet and lock on it with HMD and refine with FLIR. It looked good for a while but then I started seeing discrepancies in altitude after only 10 min of flying ( lots of turning and high G though) The mission was set for 1985. I haven't been very successful with INS updates. I'll play with it some more but form what IvanK and others are saying... it won't work too well.  

 

Here's a short vid from the dive deliveries. (1985 mission - before alignment degraded). No wind.

https://youtu.be/3GQBeAKabt4

 

 

 

Link to comment
Share on other sites

On 7/29/2021 at 8:03 PM, Gripes323 said:

 

TBH, I'm not quite sure what changed since then but currently I'm able to whack the target in AUTO or CCIP regardless of target altitude, as long as I'm in a dive (I only tried between 15 and 20 deg. dive at M .8 )  Level deliveries in auto are off, mostly short and left (3 out of 4)  I used 1 MK83 on station 5 (center) with ATFLIR on a cheek.

I'm gonna do some more tests today with HMD target acquisition and FLIR.  So far I tried to set up a target at 10k feet and lock on it with HMD and refine with FLIR. It looked good for a while but then I started seeing discrepancies in altitude after only 10 min of flying ( lots of turning and high G though) The mission was set for 1985. I haven't been very successful with INS updates. I'll play with it some more but form what IvanK and others are saying... it won't work too well.  

 

Here's a short vid from the dive deliveries. (1985 mission - before alignment degraded). No wind.

https://youtu.be/3GQBeAKabt4

 

 

 

Ins degradation shouldnt affect bomb solution calculation. Unless you use pre entered coordinate, then the ins will drift yes (though not implemented yet)

like pointed out before, with no IFA. Cues provided in CCIP or Auto are wrong as demonstrated by Ivan

I hope ED read this and will fix it , maybe when they implement a proper ins

Link to comment
Share on other sites

Ins degradation shouldnt affect bomb solution calculation. Unless you use pre entered coordinate, then the ins will drift yes (though not implemented yet)

like pointed out before, with no IFA. Cues provided in CCIP or Auto are wrong as demonstrated by Ivan

I hope ED read this and will fix it , maybe when they implement a proper ins
The drift is definitely modeled, actually *I we modeled* when in NAV.

Mobius708

Link to comment
Share on other sites

12 hours ago, Hulkbust44 said:

The drift is definitely modeled, actually *I we modeled* when in NAV.

Mobius708
 

You sure about that? Ed specifically said they intent to model it along with the new INS for the hornet never saw it

And from my testing my waypoint are always dead accurate , no drift . (yes tested without IFA)

Link to comment
Share on other sites

On 7/28/2021 at 11:41 AM, Harker said:

Except we don't have DTED. AFAIK it came as part of the TAMMAC upgrade, which we should have, but we don't.

The DCS A-10C, however, indeed uses INS/GPS, attitude and DTED.

So ED should either implement TAMMAC or should remove the DTED functionality we appear to have out of nowhere.
 

It's been discussed quite a lot on the forums and on the Discord. The general consensus is that given the features we have and the specific model/time, we do have TAMMAC. We should actually get the updated TAMMAC HSI, which we don't.

Link to comment
Share on other sites

31 minutes ago, toilet2000 said:

It's been discussed quite a lot on the forums and on the Discord. The general consensus is that given the features we have and the specific model/time, we do have TAMMAC. We should actually get the updated TAMMAC HSI, which we don't.

Yes, we should have TAMMAC, but as you say, our HSI layout is pre-TAMMAC. So, until the HSI changes to reflect the inclusion of TAMMAC, I assume we do not have it.

 

If we're supposed to have TAMMAC, according to ED, then they need to change the HSI.


Edited by Harker

The vCVW-17 is looking for Hornet and Tomcat pilots and RIOs. Join the vCVW-17 Discord.

1197644828_Screen_200911_044202-Copy.png.74d8c09ee9060cffd7408a75ab2c13ef.png

HP Reverb G2, Z370 Aorus Gaming 7, i7-8700K, 3090 FTW3 Ultra, 32GB DDR4, 960 Pro, 970 Evo Plus, WD Gold 6TB, Seasonic Prime Platinum

F/A-18C, AV-8B, F-16C, JF-17, A-10C (C II), M-2000C, F-14, BS2, UH-1H, P-51D, Sptifire, FC3

Link to comment
Share on other sites

Yes, we should have TAMMAC, but as you say, our HSI layout is pre-TAMMAC. So, until the HSI changes to reflect the inclusion of TAMMAC, I assume we do not have it.
 
If we're supposed to have TAMMAC, according to ED, then they need to change the HSI.
Yet TAMMAC is requirement for the TAWS and subsequent systems (DTED, DMS) that we *do* have.
@BIGNEWY

Note for ED: see 2.17.4 in A1-F18AC-NFM-000

1-2-145

Mobius708

Link to comment
Share on other sites

9 hours ago, Harker said:

Yes, we should have TAMMAC, but as you say, our HSI layout is pre-TAMMAC. So, until the HSI changes to reflect the inclusion of TAMMAC, I assume we do not have it.

 

If we're supposed to have TAMMAC, according to ED, then they need to change the HSI.

 

The updated HSI is the only thing we don't have from the TAMMAC upgrade. So I assume we simply don't have the right HSI and we have TAMMAC, which actually fits the CCIP behavior.

Link to comment
Share on other sites

9 hours ago, toilet2000 said:

The updated HSI is the only thing we don't have from the TAMMAC upgrade. So I assume we simply don't have the right HSI and we have TAMMAC, which actually fits the CCIP behavior.

"....which actually fits the CCIP behavior"

 

What do you mean by this statement ? exactly what behavior ?

Link to comment
Share on other sites

16 hours ago, IvanK said:

"....which actually fits the CCIP behavior"

 

What do you mean by this statement ? exactly what behavior ?

INS position influencing the CCIP solution. There are some issues (as said in one of my previous comment), but with DTED that came with TAMMAC, INS/GPS position should change the CCIP solution calculation, as it would not only rely on current radar altitude when no AGR/TPOD ranging is preset, but rather on position, attitude and the intersection of the weapons predicted path with the elevation map given the current INS/GPS position.

Link to comment
Share on other sites

14 minutes ago, toilet2000 said:

INS position influencing the CCIP solution. There are some issues (as said in one of my previous comment), but with DTED that came with TAMMAC, INS/GPS position should change the CCIP solution calculation, as it would not only rely on current radar altitude when no AGR/TPOD ranging is preset, but rather on position, attitude and the intersection of the weapons predicted path with the elevation map given the current INS/GPS position.

 

Did you read the thread from the start? What ever features you are talking about has nothing to do with what this thread is about.

[sIGPIC][/sIGPIC]

Win10 64, Asus Maximus VIII Formula, i5 6600K, Geforce 980 GTX Ti, 32 GB Ram, Samsung EVO SSD.

Link to comment
Share on other sites

27 minutes ago, Svend_Dellepude said:

 

Did you read the thread from the start? What ever features you are talking about has nothing to do with what this thread is about.

Did you read my first comment in your thread? 



What ever feature I am talking about has definitely a link with the thread and an issue with it actually makes the CCIP unusable with INS drift (because AGR doesn't override the DTED range estimation).

Moreover, I was answering question others asked me.

Link to comment
Share on other sites

I did read your comment and to be honest i don't think the issue is as deep as you make it up to be. The problem is basically that the + is tied to the drift of the INS, which it shouldn't be. 

Maybe has something to do with how F-18 calculated ballistics before INS was a thing or simply a bug. When you lock things up with the radar, the drift is gone and you now get the correct position on the ground, which shouldn't happen because AFAIK the radar only knows it's position from the INS. 

[sIGPIC][/sIGPIC]

Win10 64, Asus Maximus VIII Formula, i5 6600K, Geforce 980 GTX Ti, 32 GB Ram, Samsung EVO SSD.

Link to comment
Share on other sites

1 hour ago, Svend_Dellepude said:

I did read your comment and to be honest i don't think the issue is as deep as you make it up to be. The problem is basically that the + is tied to the drift of the INS, which it shouldn't be. 

Maybe has something to do with how F-18 calculated ballistics before INS was a thing or simply a bug. When you lock things up with the radar, the drift is gone and you now get the correct position on the ground, which shouldn't happen because AFAIK the radar only knows it's position from the INS. 

You really don't understand how the CCIP cross is computed.

I suggest looking a bit a the tutorials on the A-10C, as it only uses it's INS/GPS combined with DTED (Digital Terrain Elevation Database) to compute the slant range and point of impact.

Old Hornet models used to simply compute the CCIP slant range (and thus the cross position) using the radalt or baro altitude (like the Mi-24) or using the AGR when commanded to (Sensor Select to HUD). The radar doesn't directly need any INS data to compute slant range.

Hornets that received the TAMMAC upgrade (like ours, except for the HSI) have the integration of the DTED, meaning the aircraft, knowing its position from the INS/GPS, can infer the slant range by correlating its slant angle and INS position with the elevation database and finding the intersection of both. This gives slant range, but only if the INS is accurate. This is generally not an issue because moments without GPS coverage are often small enough that the INS won't drift enough for the accuracy to be that much lower.

The issue arises when you completely remove the GPS on a whole mission with a bird made to fly with GPS and with the DTED integration.

TL;DR Yes, the CCIP cross should depend on the INS position when not in AGR mode.


Edited by toilet2000
  • Like 1
Link to comment
Share on other sites

We are talking about "Old Hornet models"  pre GPS Hornet, without DTED and using AGR. That is what DCS is simulating if you have a mission date before 28 March 1994 as I see it.

 

In any case regardless of Slant range source the DIL shouldnt be displaced from the vertical nil wind and in balanced flight !

 

 

  • Thanks 1
Link to comment
Share on other sites

6 hours ago, IvanK said:

In any case regardless of Slant range source the DIL shouldnt be displaced from the vertical nil wind and in balanced flight !

 

 

 

👍.  That was my understanding of how this worked. If you were flying over fairly flat terrain with no elevation changes... who cares if INS drifted 1k meters. You should be getting a good range and the DIL should point down... unless, the MC and the standby instruments didn't know which way was straight down, :confused:

Even if the drift took you from flat to over a tall hill, the cross would be off but not slanted, would it...

I need to play around with this mess and see how all of this stuff works here.


Edited by Gripes323
Link to comment
Share on other sites

Posted (edited)
5 hours ago, toilet2000 said:

You really don't understand how the CCIP cross is computed.

 

Maybe not, but If you would be so kind, to create a new mission based in 1981, place a vehicle of your choice, then grab an F-18 with some dumb bombs, fly around for minimum half an hour and then try to land a bomb on the vehicle based on the CCIP solution the FCC gives you.

 

This is not about elevation data but lateral displacement of the CCIP +.


Edited by Svend_Dellepude

[sIGPIC][/sIGPIC]

Win10 64, Asus Maximus VIII Formula, i5 6600K, Geforce 980 GTX Ti, 32 GB Ram, Samsung EVO SSD.

Link to comment
Share on other sites

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...