Future Point Defense Options

A ship with 700 gs of acceleration fired upon from 1 light second with a grazer (or a missile commiting to its attack run on the final 1 second of approach), which needs to move 200 meters to generate a miss either due to position or due to angling its sidewall/wedge to bend/deflect the shot, needs to accelerate with at least 1/35 its acceleration in any direction. Assuming vector accuracy of 1/10 of a degree, it has roughly 150 million potential dodge options.

Now imagine that the attacker is trying to guess what the dodge option will be so it can land a hit, via effectively behavioral analysis of past movements, via data gathered from sensors. At the same time, the defender is trying to spoof its exact position, exact acceleration, exact vector as well as outpredict the attacker and choose a dodge option that would be unlikely. Both attacker and defender have to account for sensor ghosts, sensor accuracy margins, relativistic distortion, physical distortion of their own electronics due to thousands of gs of acceleration, gravitational distortion of any readings due to multiple layers of forcefields and all kinds of jamming.

And they have to play this kind of future-prediction-chess in mere seconds without human interference. I'd call this a game with 2-3 orders of magnitude more "pieces" than chess, in a board with 6 orders of magnitude more options, played in 3 orders of magnitude less time. Ergo, Honorverse tactical computers need 12 orders of magnitude more computing power than real-life computers that can play chess.




Just saying...

Belial666 wrote:A ship with 700 gs of acceleration fired upon from 1 light second with a grazer (or a missile commiting to its attack run on the final 1 second of approach), which needs to move 200 meters to generate a miss either due to position or due to angling its sidewall/wedge to bend/deflect the shot, needs to accelerate with at least 1/35 its acceleration in any direction. Assuming vector accuracy of 1/10 of a degree, it has roughly 150 million potential dodge options.

Now imagine that the attacker is trying to guess what the dodge option will be so it can land a hit, via effectively behavioral analysis of past movements, via data gathered from sensors. At the same time, the defender is trying to spoof its exact position, exact acceleration, exact vector as well as outpredict the attacker and choose a dodge option that would be unlikely. Both attacker and defender have to account for sensor ghosts, sensor accuracy margins, relativistic distortion, physical distortion of their own electronics due to thousands of gs of acceleration, gravitational distortion of any readings due to multiple layers of forcefields and all kinds of jamming.

And they have to play this kind of future-prediction-chess in mere seconds without human interference. I'd call this a game with 2-3 orders of magnitude more "pieces" than chess, in a board with 6 orders of magnitude more options, played in 3 orders of magnitude less time. Ergo, Honorverse tactical computers need 12 orders of magnitude more computing power than real-life computers that can play chess.




Just saying...

[Edit - Ignore this whole post. I had a thinko and misremembered energy range ]
That's part of why energy range is only 1/3rd of that. 100,000 km; not 1 lightsecond (299,792.458 kilometers)

So the short range makes it easier; plus of course wedges can change heading, acceleration, or roll instantly. So that helps you guess where the target is more likely to go.

But nobody said energy range combat was easy, especially between smaller ships that are maneuvering wildly looking for an opening.

Jonathan_S wrote:

Belial666 wrote:A ship with 700 gs of acceleration fired upon from 1 light second with a grazer (or a missile commiting to its attack run on the final 1 second of approach), which needs to move 200 meters to generate a miss either due to position or due to angling its sidewall/wedge to bend/deflect the shot, needs to accelerate with at least 1/35 its acceleration in any direction. Assuming vector accuracy of 1/10 of a degree, it has roughly 150 million potential dodge options.

Now imagine that the attacker is trying to guess what the dodge option will be so it can land a hit, via effectively behavioral analysis of past movements, via data gathered from sensors. At the same time, the defender is trying to spoof its exact position, exact acceleration, exact vector as well as outpredict the attacker and choose a dodge option that would be unlikely. Both attacker and defender have to account for sensor ghosts, sensor accuracy margins, relativistic distortion, physical distortion of their own electronics due to thousands of gs of acceleration, gravitational distortion of any readings due to multiple layers of forcefields and all kinds of jamming.

And they have to play this kind of future-prediction-chess in mere seconds without human interference. I'd call this a game with 2-3 orders of magnitude more "pieces" than chess, in a board with 6 orders of magnitude more options, played in 3 orders of magnitude less time. Ergo, Honorverse tactical computers need 12 orders of magnitude more computing power than real-life computers that can play chess.

Just saying...

That's part of why energy range is only 1/3rd of that. 100,000 km; not 1 lightsecond (299,792.458 kilometers)

So the short range makes it easier; plus of course wedges can change heading, acceleration, or roll instantly. So that helps you guess where the target is more likely to go.

But nobody said energy range combat was easy, especially between smaller ships that are maneuvering wildly looking for an opening.

umm, no. Energy weapon range is IIRC about half a million kilometers, or 1.5 light seconds (ish). At the battle of Cerebus the ESN took out virtually all of the PN task force from about 20% past that.

That said, the idea of 'dodging the bullet' to multiply the hit complexity seems a bit screwy. There's two problems, one is the law of momentum, you can't simply jink the ship out of position far enough, and second, to time the jink you'd have to know precisely when the energy weapon fired... by which time you're dead. Plus-- most likely you're not being targeted by one graser but as many as are in the broadside. Doesn't take but one or two to totally screw up your whole day.

SharkHunter wrote:

Jonathan_S wrote:That's part of why energy range is only 1/3rd of that. 100,000 km; not 1 lightsecond (299,792.458 kilometers)

So the short range makes it easier; plus of course wedges can change heading, acceleration, or roll instantly. So that helps you guess where the target is more likely to go.

But nobody said energy range combat was easy, especially between smaller ships that are maneuvering wildly looking for an opening.

umm, no. Energy weapon range is IIRC about half a million kilometers, or 1.5 light seconds (ish). At the battle of Cerebus the ESN took out virtually all of the PN task force from about 20% past that.

Mea Culpa.
I somehow got energy range and GL range mixed up. Energy range is, as you said, at least 400,000 km (more if the target doesn't have a sidewall interposed)

A ship with 700g accel only has said acceleration in one direction. Forward. RFC has his SD requiring 12! minutes to flip end of end... Spin rate is obviously faster, but its Yaw and Pitch rate for his ships is slower than a snail.

His ships are modeled on SAILING vessels. Ergo, the very slow turn radius.

Relax wrote:A ship with 700g accel only has said acceleration in one direction. Forward. RFC has his SD requiring 12! minutes to flip end of end... Spin rate is obviously faster, but its Yaw and Pitch rate for his ships is slower than a snail.

His ships are modeled on SAILING vessels. Ergo, the very slow turn radius.

http://infodump.thefifthimperium.com/en ... gton/128/1

Sorry, the 12 minute figure is only for a 90 degree turn. A full flip of a SD would take 24 minutes.

Granted, this was posted in 1998, so the SD(P)s and other new classes may be quicker. But not by much - given the tonnage creep, they may even be slower.

munroburton wrote:

Relax wrote:A ship with 700g accel only has said acceleration in one direction. Forward. RFC has his SD requiring 12! minutes to flip end of end... Spin rate is obviously faster, but its Yaw and Pitch rate for his ships is slower than a snail.

His ships are modeled on SAILING vessels. Ergo, the very slow turn radius.

http://infodump.thefifthimperium.com/en ... gton/128/1

Sorry, the 12 minute figure is only for a 90 degree turn. A full flip of a SD would take 24 minutes.

Granted, this was posted in 1998, so the SD(P)s and other new classes may be quicker. But not by much - given the tonnage creep, they may even be slower.

I wonder if the "accel" in one direction is less the issue than inertial compensation. Let's face it, turning any significant level of "Km/sec" is going to take a while, let's say you're at 30,000Km/sec earth to moon in 10 seconds, you're not going to turn in probably anything less than minutes because that 500g or so accel/decel needs time to take effect, and squishing the humans into paste precludes turning faster.

I'd be willing to wager that even an SD(p) could turn 90 or 180 degrees quite a bit more quickly from a relative stand still.

The ship is in total free fall, that is what the compensator does. So that makes no sense.

kzt wrote:The ship is in total free fall, that is what the compensator does. So that makes no sense.

The timing for a 90 degree turn in a small military jet pulling 6-7 gees (been there done that) took several seconds, for an SD we're talking something like 200,000 times more weight, but even i the IC allows the cheat of say 250-500:1g, that still leaves a helluva long turn time/radius at any appreciable fraction of C.

So my thought is that if the compensator generated a 360 degree bubbe (pure free fall) it would make no sense, but I'd assume that the IC has to be tuned and aligned with the same vector as the acceleration force (the impeller which of course is also the wedge's orientation), otherwise the gravitic cross/forces would start to get... interesting to say the least.

SharkHunter wrote:

kzt wrote:The ship is in total free fall, that is what the compensator does. So that makes no sense.

The timing for a 90 degree turn in a small military jet pulling 6-7 gees (been there done that) took several seconds, for an SD we're talking something like 200,000 times more weight, but even i the IC allows the cheat of say 250-500:1g, that still leaves a helluva long turn time/radius at any appreciable fraction of C.

So my thought is that if the compensator generated a 360 degree bubbe (pure free fall) it would make no sense, but I'd assume that the IC has to be tuned and aligned with the same vector as the acceleration force (the impeller which of course is also the wedge's orientation), otherwise the gravitic cross/forces would start to get... interesting to say the least.

I think you're confusing heading with vector.

It's the same acceleration to change heading 90 degrees no matter what your base velocity is (say, relative to the nearest star). Sure if you're going an appreciable fraction of c it'll take many hours to actually alter your vector. But your heading should be changable just as quickly as if you were at rest relative to the local system. There's no medium to "corner" against, so your speed is irrelivant to your heading change ability.


But for some reason SDs have lethargic heading changes; even before they start trying to build up a new, or counteract an old, vector.