F-14 radar gunsight solution incorrect

[some1]

1 hour ago, Temetre said:

AND, this is the tricky part, then extrapolate what happens if the planes keep turning. Which would be difficult to get accurate in the first place; but worse, if the enemies turn rate/speed/roll/acceleration isnt staying exactly the same, that extrapolation could be wildly off by itself. So for example if the enemy does a slight roll additional to its turn, the aim point would already be off again.

This report is about a scenario where the enemy flies a constant turn and you follow him with steady tracking. The enemy position is constant relative to your aircraft and the gun. No target extrapolation is needed, just an accurate bullet path / time of flight calculation.

Basically an ideal conditions, If the solution is inaccurate here, then it's inaccurate in any other situation, except for the rare occurrences when two errors cancel out, or the error is too small and you still manage to hit the target. 

In the F-4 the bigger problem is achieving a good stable radar lock. The gunsight itself seems to be more accurate in such scenario than in F-14, but I didn't do much testing recently.

[Snappy]

2 hours ago, some1 said:

This report is about a scenario where the enemy flies a constant turn and you follow him with steady tracking. The enemy position is constant relative to your aircraft and the gun. No target extrapolation is needed, just an accurate bullet path / time of flight calculation.

Basically an ideal conditions, If the solution is inaccurate here, then it's inaccurate in any other situation, except for the rare occurrences when two errors cancel out, or the error is too small and you still manage to hit the target. 

In the F-4 the bigger problem is achieving a good stable radar lock. The gunsight itself seems to be more accurate in such scenario than in F-14, but I didn't do much testing recently.

I agree with your initial OP on this . Was this ever officially acknowledged or tracked as a bug?

[Temetre]

vor 4 Stunden schrieb some1:

This report is about a scenario where the enemy flies a constant turn and you follow him with steady tracking. The enemy position is constant relative to your aircraft and the gun. No target extrapolation is needed, just an accurate bullet path / time of flight calculation.

On a basic level, you already need to extrapolate because you want bullets cross the enemies flight path in the future. 

And the enemies position isnt constant relative to the bullets that youre firing. Because the bullets fly in a straight line (+ ballistic arc), while both you and the enemy are turning. And the rate of turn/acceleration might be actively changing, so the enemies speed might be increasing/decreasing, which has to be estimated to be perfectly accurate in a turn fight.

vor 4 Stunden schrieb some1:

In the F-4 the bigger problem is achieving a good stable radar lock. The gunsight itself seems to be more accurate in such scenario than in F-14, but I didn't do much testing recently.

Admittedly I havent tested much with the F4s gunsight, but I found its lag alone makes it often hard to use in close combat.

Edited June 19 by Temetre

[some1]

2 hours ago, Temetre said:

On a basic level, you already need to extrapolate because you want bullets cross the enemies flight path in the future. 

And the enemies position isnt constant relative to the bullets that youre firing. Because the bullets fly in a straight line (+ ballistic arc), while both you and the enemy are turning.

That's the whole point, you don't have to. Gyro gunsight from WWII works without radar, not knowing anything about enemy position and velocity.

If you use your aircraft as frame of reference, then in a stable tracking shot enemy position is relatively constant. Then the whole problem can be reduced to: "where the bullets will be relative to my aircraft after a time of flight that is required to cover the distance to target" and then paint the gun cross at that position. If the pilot puts the cross on the target, he'll score a hit.

Of course solving that ballistic equation is not a trivial task, with many parameters that affect bullet trajectory, and the enemy in real life (or in simulated combat) is rarely as cooperative as AI forced to fly in circles. But you can't expect much accuracy in a more complex situation from a gunsight that doesn't work well even in a "best case" scenario.

null

Edited June 19 by some1

[Schmidtfire]

This is very likely a bug based on other DCS modules. I don't think I have ever seen a module with radar ranging guns working correctly
"out of the box". They all undershoot. Aim for the canopy and if you are lucky - hit the tail.

[Snappy]

On 6/19/2024 at 10:05 PM, Schmidtfire said:

This is very likely a bug based on other DCS modules. I don't think I have ever seen a module with radar ranging guns working correctly
"out of the box". They all undershoot. Aim for the canopy and if you are lucky - hit the tail.

Could it be a global issue based on core DCS environment?  I.e. that the actual gunsight solution calculation is correct, but things like the air density profile in DCS atmosphere differ very slightly compared to real world air atmosphere, resulting in tiny additional bullet slowdown during flight. It would only need be a tiny bit off to cause this. Or muzzle velocity being slightly low. I don't know.

But I also noticed it in other modules (Mirage F1 for example) . Its a small issue , but annoying. Would be nice to see this fixed.

Many of the real-world photos taken during training combat, showing  various aircraft in a gunsight with pipper on would result in misses if it was like it is in DCS.

Edited June 22 by Snappy