DataPacRat replied to me: > > Or the users of this fiendish weapon are really trying to > > exterminate the entire enemy population, and Our Steel-Jawed > > Heroes are the ones fighting against the Terrible Menace. > > Our Heroes were looking for a system which can identify, and /keep/ > identified, anyone who attacked them - and anyone who supported them > by, eg, storing ammo, hiding insurgents and supporting their claims to > be innocent civilians, and was otherwise someone who would be > considered part of the attackers' forces if their actions were known > in full. (Admittedly, the original fiction included a sufficient > quantity of attackers that certain target areas ended up being > essentially depopulated, at least before the non-combatants started > kicking out the combatants and surrendering...)
If you want robots to replace HUMINT '15 minutes into the future,' as you put it, you are looking for a miracle, not science fiction. "You were observed -- three times in the last 48 hours -- talking to a known contact of the man who has the same first name, shoe size, and date of birth as the local terrorist leader. That means you're an unlawful combatant, too, and you're hereby sentenced to indefinite internment without the possibility of a trial." "But, but, he is the only grocer left in the village. Everybody knows him." "Mark that grocer for intensive interrogation, then. One more case before lunch?" http://en.wikipedia.org/wiki/Shahid_Malik#Problems_at_US_airports And "non-combatants" who are able to kick out combatants must have been combatants themselves until the decided to change sides. http://www.google.com/hostednews/afp/article/ALeqM5iKXQC2HwynedmTU1YcjQQFpyPA_g > Okay - with 5 lasers and a ROF of 16 each, it looks like this tank > could plink 80 Pigeons per second before getting saturated. It might get perhaps get one pigeon with every other four-round burst, walking the bursts into each target. If all lasers can bear because the tank is being swarmed, that gives ten targets per second. > That's > about $240k of materiel lost per second - so a Pigeon attack on such a > tank would need to hide as much as they could from said tank, before > as many as possible swarmed it at once, from as close as possible. How close is possible? If the terrain allows the pigeons to sneak close without being engaged, it will also restrict them to just a few avenues of approach. > If > the Pigeon-net knew the tank was around, the HEDP self-destructs might > be swapped out for HEATs... and, of course, whichever Pigeons survive > the point-defense should try to aim for the most vulnerable points, > which appear to be the two tracks. How many HEATs does the net have in stock? And who controls the net if tank lobs a few HE shells into the base? > If a flock of 100 attacked such a tank from cover... then 20 would > make it through the point-defense, and if they had HEAT warheads, the > tracks' armor would be effectively 8, allowing 272 damage through per > Pigeon, or about 2720 per track. So, for $300k, a $5.7M tank gets > immobilized, to get swatted by artillery at leisure. (If nearby cover > is unavailable - then the attacking swarm would have to be large > enough to have some survive the trip from wherever they are hiding to > make it to the tank, for a corresponding increase in cost.) > > How'm I doing so far? Lining up all your pigeons in a row for a few beehives, I'd say. Like many enthusiasts for a neat new idea, you take a best case for one side and a worst case for the other side. > Hrmph. And I was so looking forward to having this program be a > primary spur to help jump-start the solar-satellite program, which > would be useful as the basis for having actual manufacturing > infrastructure in space... maybe I'll fall back on more general > military energy needs, such as air conditioning. Separate the information gathering role from the attack role. Beamed power from orbit to the intel base on the ground, and beamed power from the intel base to sensor UAVs circling above the ceiling of light air defenses (which puts them into line-of-sight to the base, too). The UAV operators on the ground watch the poor bloody infantry live and in color as they slog through the ruins, and call in helpful advice. > > A C1 robot brain would have IQ 4. Animals with that level of > > intelligence don't usually bash their heads in. > > Then is the CompNav program basically pointless for robots? Computer navigation allows a robot to deliver a pizza to Park Avenue 123, in minimum time and without going the wrong direction in an one-way street (if that is noted in the maps), and if there are enough terrain features it allows the robot to navigate without GPS, by looking for landmarks. Raw IQ in the 4-6 range allows it to figure out how to get around an unexpected parked car where there was no car last time. > Step 1: Locate anyone who shoots at the Pigeons, or other allied forces. > Step 2: Keep track of them. Follow them and identify their > interactions with others, such as ammo-bearers or civilian 'covers'. > Step 3: Profit! by blowing 'em up. See above for pitfalls with that scheme. As soon as the locals figure that pigeons will come after them no matter what they do, they'll fight back. > I'm less worried about launching all the Pigeons from a container at > once than simply shipping them around - I'd be perfectly happy to send > them as cargo, if that was legal. Yes. Consider the notes re cargo ramps. > Of course, the Pigeons could be lined up on top of each other, with a > spring-loaded base, like a pistol's magazine, with bunches of them > popping out of each individual hatch. :) There is a precedent in GURPS Traveller to fire vehicles from missile tubes. > Hm... I was hoping that when deployed that way, the first container on > the spot would be one full of Pigeons, and could start launching right > away, with the other containers adding their various supports as they > arrived. Earlier you explained how all sorts of complicated analysis gets handled by the base, and now you want to launch a flock in autonomous mode? Regards, Onno _______________________________________________ GurpsNet-L mailing list <[email protected]> http://mail.sjgames.com/mailman/listinfo/gurpsnet-l
