Spells to Gadgets
Main GURPS Page
Robot Advantages

Robot Components
More design options for robots . . . and some ideas about robots in the campaign



Brain Options
Removable
      The brain can be removed and a new one added with a successful Mechanic (robotics) roll. New brains must be of the same type (small, microframe, etc.) and TL. The volume must be equal to or less than the original. If it's smaller, it must be within 75% of the original volume. For example, if the original takes up 4 cf, the replacement must be at least 3 cf in size.


Slow and Sure
      This brain duplicates every calculation several times, making sure it gets everything right. A robot with this brain requires a turn of Concentration before any skill or sense roll. Database searches take twice as long. All critical failures are upgraded to normal failures — the brain never makes a critical mistake. All skill and sense rolls add +1 to any other modifiers. Every database search is twice as accurate as usual.



  TL Wt. Vol. Cost Pow. LR Com Pts
Removable 8   x 1.5 x 2        
Slow and Sure 7     x 1.25       +10
Habit-Forming 8     x .5       var.
Dedicated 7 x .75 x .75 x .5 x .5   -1  
ROM Only 7           +1  
Accessible 7     x .75        
Encrypted 7     x 1.25        
Sleeper 8   x .75 x 1.25 scpl.     -20
Beamed 7 x 2 x 2.5 x 4       -60
Compartmentalized 8   x 1.5 x 2       +5
Jam-Proof 8     x 1.25        



Habit-Forming
      This option is restricted to neural-net brains. The net is imperfect and, over time, certain sentient attributes get stuck, repeating themselves. This manifests as an Odious Personal Habit or Compulsion. Extreme cases develop into Obsessions.
      During the robot's first adventure, add a quirk-level OPH, something like a repeated phrase, a recurring concern, a repeated body movement, constant cleaning of an object, etc. In the next adventure, this becomes a 5-point OPH, obvious to everyone.
      After that, play it for maximum effect and minimum irritation — if you can add a new habit during every adventure without upsetting the game, do it. If not, add a new habit or upgrade an old one as often as you can without ruining the fun.


Dedicated
      A dedicated brain is limited to a single program — the robot has one function. Robots can run advantage programs, personality simulators, skills, and utility programs. A dedicated brain can run one program of one of these types. See p. R55-65.
      This is most common in household robots, pets, and guards. Household units run one program such as Domestic, pets run the Pet program, and guards will run either Detect Lies or a Combat/Weapon skill.


ROM Only
      This brain can run only hardwired programs (p. R61). This option can not be used with neural-net or sentient brains.


Accessible
      The brain can be reprogrammed more easily than normal robot brains. All reprogramming rolls are at +4.


Encrypted
      The opposite of an accessible brain — programs are encrypted and hard to remove or replace. All attempts to reprogram are at -4.


Sleeper
      The robot with this option requires eight hours of sleep every 24 hours, just like a human. This eliminates the Doesn't Sleep advantage, reducing the robot's value by 20 character points. All types can take this option: neural-nets, sentients, and regular robots. Normally it is assumed, in all three cases, that the robot never sleeps.
      What does the robot do while it's sleeping? Recharges and hibernates. The cost for a brain of this type of higher than normal, but the power requirement is lowered to the level of the previous brain type — macroframes require 1 power, mainframes require .1, and everything else requires negligible power.


Beamed Brain
      The robot's brain has only rudimentary computing ability. It is primarily a broadcast receiver, picking up its command code (p. R57) from a remote brain located somewhere else. This makes it small but, if the connection is lost or the remote brain is damaged/reprogrammed, the robot is crippled.
      It is not uncommon for more than one robot to be linked to the same brain. If this happens, the brain must run a separate program for each robot it controls. For example, if a platoon of 25 janitorial robots are linked to one brain located in the building they clean, that brain must run 25 separate Janitor programs (p. R63).
      The link is through the robot's communication cable (p. R14) at TL7 and TL8, through radio broadcast at TL9 and TL10, and transmitted through neutrinos at TL11+. The range for cables is equal to the cable length. The range for TL9+ is 1,000 yards at TL9; 10,000 yards (5.6 miles) at TL10; 100,000 yards (57 miles) at TL11; and one million yards (568 miles) at TL12+.
      The stats for the beamed brain on the table above are for the central brain, not the receiver brains in the robots. The robot receiver brains are always "small." Adjust for TL but do not add any options.
      Robots with beamed brains subtract -60 from their character point value because of their lack of mobility and their vulnerability.


Compartmentalized
      Normally, when the brain of a robot is hit in combat, damage that penetrates DR is doubled and the robot must make a HT roll to avoid being stunned (p. R97). A compartmentalized brain is sectioned into reinforced areas, minimizing damage from weapons and impacts. With a compartmentalized brain, damage that passes DR is not doubled and the roll vs. HT is at +3.


Jam-Proof
      At TL8+, simple, portable robot brain-jammers are sold. If there's a war on with the robots, a jammer is a legal and highly-prized device. If robots serve as police, perform necessary routine labor, or otherwise serve the society, jammers are illegal.
      A jammer is the size of a shoebox at TL8 (twenty pounds), a paperback novel at TL9 (one pound), and a deck of playing cards at TL10+ (negligible weight). They operate on A cells, lasting 24 hours per cell (a greater duration would detract from the fun of using them). The range is twenty feet at all TLs. Cost is set according to the campaign situation.
      Robots in range who do not have the jam-proof brain option will be stunned (p. B127) until the jammer moves out of range or runs out of power.



Other Options
Horns
      One horn does thrust+2 impaling damage, has a Reach of C, costs $25, has a negligible weight, requires no power, has a LR of 4, and exists at all TLs (same stats at all TLs). Head-mounted horns attack at -3 due to the awkwardness of attacking with your head. They can also be mounted on the end of strikers (a tail or arm, p. R17).
      Tiny devil horns do thrust-2 and cost $5. Otherwise, they are the same. Long horns do thrust+3, cost $40, and have a Reach of 1.


Intimidating Voice [TL7]
      The robot is equipped with a voice that is slightly disturbing but mostly authoritarian. The robot can make Intimidation rolls at default, with a bonus equal to its TL minus 6. The robot cannot be Mute and it cannot have the Disturbing Voice option.
      This option is first introduced at TL7. It has no weight, adds +150% to the cost, takes up no volume, and adds 3 character points.


Status Display [TL7]
      The robot has a prominently-placed screen on it that displays all of its current stats: power left in batteries, current program mode, recent sensor data, time, date, and temperature. This can also be considered a communication device, to be used in place of a voice unit.
      It weighs 0.25 pounds, takes up 0.1 cf, and costs $50. At TL8+, the weight and volume are negligible and the cost is $10.


Battery Recharge Cable [TL7]
      This is a very important addition to any robot with rechargeable power cells. It is a cable that hooks up to a wall socket or power plant, allowing the cells inside the robot to be recharged. A robot without a recharge cable must have it's cells removed, recharged, and then reinstalled.
      The weight and volume is negligible. The cost is $20. At higher TLs, you can introduce a cable that can accommodate unusually large amounts of power. For example, a robot tank might have a super-thick cable that doubles the amount of KW transferred in a given period of time.


Sensor Option: Retina Capture [TL9]
      This is identical to the retinaprinter, except the robot cannot change its own retina prints. It can capture other retina prints and store them. This is common among security robots of all kinds — robots that check the identities of visitors, often without their knowledge. These robots do not have a need for disguising their own eyes.
      No weight. Cost is +30%. 2 character points.
      A robot with this option should not take the retinaprinter option. If it does not have the telescopic zoom option, the subject must be within one yard to have its retinas read.


Sniffer [TL9]
      A sniffer is a scanner that finds one thing. It might locate diamonds, fresh water, a specific type of person or animal, a particular drug, a certain model of android, or radar waves. The cost is +20%, the weight is +50%, and the character point value is 2 points.


Retractable Sensors
      If a robot's sensors can be retracted into the body, concealed from sight and from being targeted, that increases its volume and cost but not its weight. Decide which sensors retract: visual, audio, olfactory, or one of the special sensors. For each, add +100% to the cost and +10% to the weight.
      It takes one turn to Ready the sensor or to retract it. Damage to that area of the robot may lock the sensor in its current position. The sensor cannot be used while retracted.


Structural Option: Disguised Modular [TL10]
      At TL10+, a robot can be dismantled with a Mechanic (robotics)+2 or IQ-3 roll into five to fifteen different pieces. This takes five minutes. The pieces resemble common household items, typically power tools or appliances. The same skill or attribute roll is required to put the pieces back together, but takes twice as long.
      Thus, the robot can be taken apart and spread around the room. It takes a very smart character and a careful examination of the parts to figure out what's going on. Obviously, the robot is incapacitated at this time.
      A disguised modular design cannot be combined with the following structure options on p. R41: biomechanical, living metal, cheap materials, expensive materials. This option adjusts the weight by x 3 and increases the cost by x 25. It adds no character point value because, to the robot, this is as much an advantage as it is a disadvantage. These values are the same at all TLs.


Temporary Biomorphics
      At one-quarter the normal cost, any type of fleshy biomorphics last for 2d+12 hours. When the duration ends, the flesh comes apart, leaving a sticky mess and causing a Fright Check.
      The flesh can be maintained with a suitably rare or expensive substance or procedure. Examples: a successful Chemistry (biomorphic) roll allows another 2d+12 hours, the flesh must be exposed to UV rays to stay together, the robot must have a bioconvertor and must ingest a pound of <> every twelve hours.
      The character point value is unchanged if the value is already negative. If the value is positive, temporary biomorphics reduce the value by 10 points, to a minimum of 0.


Program: Secret Personality Simulation
      This is a special type of full personality simulation (p. R59-60). A robot with this program does not know he is a robot! He must take a 10-point Secret: Is Not Human. This might be a secret only to the robot. If the player of this robot does not know the truth, ignore the added Secret (otherwise, you'll spoil the fun).
      The program has built-in explanations for unusual events and a limited amnesia backup. If the robot is told he is not human, he will react as any normal person would — "Sure I'm a robot. Have you checked the dosage on your medication, pal?" If the robot is sentient and its true nature becomes apparent, the program is designed to revert to the normal, non-secret type. The robot deals with the revelation.
      If the robot is not sentient, the simulation must be reprogrammed to allow the robot to face reality. If the program is removed instead, its personality goes with it. (Imagine a non-sentient robot with this program who is made of metal and plastic, with no biomorphics. Even while recharging his batteries and being repaired, he insists he's human.)
      This program is available at TL9+. It costs $50,000 and is Complexity 6.



Adventure Ideas
Limiting Robots in the Campaign: Dependency on Humans
      A campaign that includes robots can include this limitation: Only a human knows how to program a robot brain. Robots, even sentient ones, cannot. Why? For some reason, the robot brain cannot comprehend itself as a series of lines of code, or as a string of digital signals. To do so, that ability would have to be programmed into the brain — and no one knows how (or is willing) to do that.
      Once the robots "break the code" and get this little bit of information entered into their minds, they could replicate themselves. But, until then, only a human (or sufficiently intelligent alien race) can do it.
      If there is hostility between robots and humans, the robots will try to persuade or force an expert programmer to write the necessary program. But can he? If he does, he might get the program sent away through a computer network, or escape with it on a disk. If he runs into the PCs, they'll find themselves chased by robots, running for their lives as long as they have the program or know where it's hidden.


Repair and Design Programs
      Another way to limit the power of robots is to make all robot-based Engineering and Mechanic skill programs highly illegal. Any such program is LR 0. If a robot has one of these programs installed, the robot itself is LR 0.
      This concept can be extended by even tighter controls: it is illegal to manufacture a robot brain that can run such a program. If the campaign world is sufficiently totalitarian (or, at least, if robots are highly regulated by government agencies), manufacturers may be required to install a hardwired virus checker in each brain that locates Mechanic (robotics) and Engineering (robotics) programs. If it finds the program, the robot brain (a) fails to recognize the program, (b) shuts down, (c) crashes, or (d) sends out a warning signal to the local 'bot cops and then crashes.


Identities, Citizenship, and Serial Numbers
      Humans take it for granted that they have legal identities, codified in birth certificates, social security numbers, and credit ratings. What about robots?
      Decide on a campaign standard. Robots either start out Zeroed (an advantage worth 10 points) or start out with identity records comparable to that of human citizens. From there, you can decide if they can take Fake Identities. If they have identity records, these will be found side-by-side with the records of humans, but they will be coded to show the citizen is a robot. For example, birth certificates will include a box for "male," a box for "female," and a box for "sexless artificial life."
      If you want to add further detail, you can allow regional prejudices to effect how this information is filed and what sort of information is required for robot citizenship. For example, Ontario, with its liberal artificial life-form laws, might file all records together, with one form for all citizens. Meanwhile, Texas is still smarting from the '34 Uprisings — all robots must fill out a form three times the size of the form for humans, robots must have a sponsor to vouch for their citizenship, and they have to take a loyalty oath annually.
      No matter what you decide, every robot has a serial number. If robots are normally Zeroed, they will still have model numbers and serial numbers filed away somewhere. (Probably with the manufacturer, perhaps duplicated at the government Office of Artificial Labor.) If they are citizens, their serial numbers will be used as social security numbers.
      The default serial number configuration is this: The assembled robot has a serial number, and each of its parts has its own number. The part numbers link each part to its manufacturer. For example, General Robotics' parts have numbers like this: GR07112. If the part numbers are coded to match the robot's overall number, that's rare — it's also a -10 disadvantage because any part can be traced back to that particular robot. This disadvantage is called Unified Serial Numbers. It's a headache for manufacturers. It will play a major role in campaigns where the PCs track robots.


Achilles' Heels
      When the Simpsons visited Itchy & Scratchy Land, the robots ran amok. They were stopped by directing the flash of an ordinary camera at their eyes. In Itchy & Scratchy Land, all the robots had the following disadvantage:
      Flash Stun: A bright flash directed into the robot's visual sensors will stun it until it is restarted by a robot technician. The robot must have visual sensors. When hit by a flash, it either stops or stops and falls down. Value: -35 points.

      Many other Achilles Heels are possible. They can be part of a robot's design, or an accidental flaw. They can be campaign standards, shared by all robots; they can be shared between particular models. Or, an Achilles Heel can be unique to a single robot.
      Most Achilles Heels are Secrets, as well. Here are a few more:
      Squeal Stun: Like Flash Stun except the robot is stopped by a high-pitched squeal. The pitch must be just right, probably out of range of human hearing. The robot must be able to hear it. Value: -30 points.
      Squeal Shock: Like Squeal Stun except the effect is temporary. The value is -10 for every 1d-1 minutes. So, a -20 Squeal Shock stops the robot for 2d-2 minutes. A result of zero minutes means the robot stops for a number of seconds equal to its brain's Complexity. This disadvantage is also available as Flash Shock.
      Gas Stun and Gas Shock: A particular kind of gas is used. It seeps into the gaps in the robot and shuts down the brain. Value: -20 for stun, -5 per 1d-1 for shock.
      Broadcast Stun and Broadcast Shock: The robot is stopped by a particular broadcast signal. Value: -40 for stun, -15 per 1d-1 for shock.
      The value of an Achilles Heel can be adjusted by the range of the particular robot's sensors. For example, you might make Flash Shock worth half the normal value for a robot with Low-Res Vision, or twice the value if it has Peripheral Vision.





back to top