Robot Brain for Turret Control

[Arkathan]

Page 102 of the Robot Handbook has this, which makes it seem like the Ship's Brain does not need to run 'Virtual Crew'.

I meant to type Virtual Crew plus one (got distracted by the better half), but now that Geir has chimed in, saying it or editing it would be cheesy. If I hadn't had to hit send quickly, I'd have included the multiple action rule adding 2 to the difficulty of every additional action, without multiple virtual crew.

 

[Geir]

A Ship's Brain is in addition to as opposed instead of a ship's computer, so the computer could run the Virtual Gunner or Crew or Auto-repair functions while the Brain was doing other things.

Couple of random thoughts that are not explicitly in the book, but perhaps should have been:

Nothing in the rulebook says you can't use the robot's bandwidth to run standard computer packages instead of or in addition to skill packages - the bandwidth can't be used to do both, but if you have the extra bandwidth sitting around after giving the robot all the skills it needs, then you could use the rest to run other programs. For instance, if you wanted to also run Virtual Crew/0 on a Ship's Brain and designed it so it had five spare bandwidth(s?) available, you could run Virtual Crew/0 in addition to its other programs and tasks.

Nothing in the rulebook limits slotted options to those listed. If you want to robot to carry some piece of equipment as an integral part of its body, then assume 1 slot for every 3 kg (or 3 litres, whichever is worse) and round up. For instance, if you had two spare slots, you could instead install up to 6 kg (or litres) of equipment and assume that the robot could power the equipment if it needs to.

 

[Vormaerin]

Page 102 of the Robot Handbook has this, which makes it seem like the Ship's Brain does not need to run 'Virtual Crew'.

Yeah, but it doesn't say it can do multiple things at once. That's what Virtual Crew or Virtual Gunner allow the ship's computer to do. I would certainly intrepret it to mean it can "man" any console it has the skill for as if Virtual Crew. But only one at a time. Or, possibly with the multitasking penalty if you want it to be more flexible.

 

[J. L. Brown]

Yeah, but it doesn't say it can do multiple things at once. That's what Virtual Crew or Virtual Gunner allow the ship's computer to do. I would certainly intrepret it to mean it can "man" any console it has the skill for as if Virtual Crew. But only one at a time. Or, possibly with the multitasking penalty if you want it to be more flexible.

There are some things which should not require a Ship's Brain to expend an action, though. If a player asks "What does my character see and hear?" than that doesn't take an action -- and a Ship's Brain should not need to take an action to get an answer to "What do the sensors show?". And while it makes sense for a character to split their minor actions up between walking (or running) and attacking (which often assumes a "stop and bring weapon to bear" step), a ship engines & weapon mounts don't require the same sort of attention so it would seem to be a little weird to rule "Well, since you have shot a couple of weapons you cannot maneuver". Talking and barking out orders is something else characters often do essentially for free -- so is a Ship's Brain expending an action to use a communications console?

 

[Condottiere]

Centralization of fire control requires computer bandwidth.

I would assume that implanting a robot brain in each weapon system is decentralization, and has no discernible bandwidth requirement.

 

[Vormaerin]

There are some things which should not require a Ship's Brain to expend an action, though. If a player asks "What does my character see and hear?" than that doesn't take an action -- and a Ship's Brain should not need to take an action to get an answer to "What do the sensors show?". And while it makes sense for a character to split their minor actions up between walking (or running) and attacking (which often assumes a "stop and bring weapon to bear" step), a ship engines & weapon mounts don't require the same sort of attention so it would seem to be a little weird to rule "Well, since you have shot a couple of weapons you cannot maneuver". Talking and barking out orders is something else characters often do essentially for free -- so is a Ship's Brain expending an action to use a communications console?

None of those things require "actions" by the character manning that station or even require anyone to be manning that station. Actions are things like Electronic Warfare or putting the engines in Overdrive. You don't have to have someone doing the "engineer" job to generate Thrust. Firing a Turret weapon does require an "action". Hailing Frequencies Open is not an "action".

It is a Pilot Action to allocate the Thrust to various different purposes, but the rules already allow the Pilot to fire a single Turret as part of their action (at -2), so the "move and shoot" is the default option for a Ship's Brain as well.

 

[Vormaerin]

To be clear, the pilot of a scout/courier can run the entire ship from the pilot's station. The only time "stations" matter is during combat when you need simultaneous roles being performed. A Ship's Brain can fill one "role" during a space combat turn, same as a player character. Unless it's computer has Virtual Crew or Virtual Gunner to extend its capabilities.

 

[J. L. Brown]

Centralization of fire control requires computer bandwidth.

I would assume that implanting a robot brain in each weapon system is decentralization, and has no discernible bandwidth requirement.

But what you describe is not the Ship's Brain; it is installing separate gunner-droids. That's allowed, but it is not the same. Similarly, it is possible to install a Ship's Brain with no actual connection to the ship's systems -- it just uses Avatars or Drones to do the jobs that sophonts would usually do. And that is pretty easy; each avatar and/or drone is essentially a character, and they get three minor (or one minor and one major) actions in a turn.

The question I had was about a ship's brain which does have (with or without the haptics upgrade) integration with the ship's systems. It is not clear what it can or cannot do in a single round; whether it gets only one action, the same number of actions as a character, or runs everything in the ship at the same time. Nor is it clear whether (or how) the TL of the Ship's Brain affects the number of tasks or actions per round. The rule on Robot's Handbook page 102:

A ship’s brain can use skill packages to perform any task a crew member can perform from a console station...

seems to imply that the limit is bandwidth -- running Expert (Powerplant Engineering) allows (routine, powerplant related engineering tasks) an action each round; and running multiple skill packages allows multiple (unpenalized) actions. But it seems like that reading is not universal.

 

[MasterGwydion]

But what you describe is not the Ship's Brain; it is installing separate gunner-droids. That's allowed, but it is not the same. Similarly, it is possible to install a Ship's Brain with no actual connection to the ship's systems -- it just uses Avatars or Drones to do the jobs that sophonts would usually do. And that is pretty easy; each avatar and/or drone is essentially a character, and they get three minor (or one minor and one major) actions in a turn.

The question I had was about a ship's brain which does have (with or without the haptics upgrade) integration with the ship's systems. It is not clear what it can or cannot do in a single round; whether it gets only one action, the same number of actions as a character, or runs everything in the ship at the same time. Nor is it clear whether (or how) the TL of the Ship's Brain affects the number of tasks or actions per round. The rule on Robot's Handbook page 102:

seems to imply that the limit is bandwidth -- running Expert (Powerplant Engineering) allows (routine, powerplant related engineering tasks) an action each round; and running multiple skill packages allows multiple (unpenalized) actions. But it seems like that reading is not universal.

My take on it, and I understand that this is not supported by the rules except perhaps tangentially, is this, for a computer, it's bandwidth is the number of "actions" (ie. available space to use programs) that can be performed in a round. A computer can use all of it's bandwidth at one time. That is part of the rules. Where My interpretation may differ is, that I run it that if it has the bandwidth to run the program, then it can take the action. With no penalty, since nowhere does it state that computers take negatives for using all of their bandwidth. Using skills like Gunnery, it has it's own rules for what and how often and with what penalties it can be used. Using a skill like Virtual Crew or Virtual Gunner, it functions more like a Commanding Officer as opposed to acting as the gunner or the engineer, and is therefore not using the rules for Gunnery.

Although, I do wish that there was a bit more clarity in the Robots book with exactly how computers/brains function so I don't have to make it up as I go along.

P.S.- What is the difference between a Computer/3 and a Brain with a base Bandwidth of 3? It is software? Hardware? Are they interchangeable?

 

[Vormaerin]

The description of the Alpha Brain states that it can simultaneously operate up to 4 consoles but takes the multitasking penalties for doing so.

The description of "Brian" states that it can handle up to 8 simultaneous tasks, again with a penalty for multitasking.

The Omega Brain also has limits, but it is less clear exactly how they work as it refers to operating drones and avatars rather than directly operating the consoles. It also doesn't state one way or another whether it takes the multitasking penalties. I would tend to think that it does, because that's the default rule and the ship's brain rules don't state an exception. YMMV.

Essentially, a ship's brain acts like a character. It can do more than one thing, but at a penalty. So, sure, your Brian can handle 8 combat actions, but it'll be at such a large penalty as to be completely useless at doing so.

 

[J. L. Brown]

The description of the Alpha Brain states that it can simultaneously operate up to 4 consoles but takes the multitasking penalties for doing so.

The description of "Brian" states that it can handle up to 8 simultaneous tasks, again with a penalty for multitasking.

Neither of those examples is particularly clear or compelling; both of the example brains have Robotic Drone Controllers installed, and the rules for simultaneous tasks & penalties are exactly what is expected from the version of the version of the drone controller which is installed. But a robotic drone controller is not required for a Ship's Brain to function. The description for 'Brian' adds to the confusion:

When interfaced with a ship, Brian is capable of flying the ship, operating bridge sensor and comm stations, firing turrets and initiating standard engineering operations.

Using the drone controller multiplexer, Brian can perform up to eight simultaneous tasks. Each drone controlled counts as a task, as does each skill function, but every additional simultaneous task beyond the first decreases checks for every task by DM-1.

So what are Brian's capabilities when NOT using the RDC? And a non-ship's brain robot can easily mount six separate RDCs, and simultaneously operate six different drones with no penalties at all -- the penalties seem to come from using a single RDC to control more than one drone each.

 

[Vormaerin]

So basically everyone, including the book's author, has said that you treat it like a character. It gets 1 action and can multitask like a character does. The rules also state that is how it works. Even if you have multiple RDCs, you are multitasking for using each separate one if doing so simultaneously, so that's six of one, half dozen of the other.

But if that isn't good enough for you, then make up your own answer. If you want it to be a fully automated starship, do that.

 

[J. L. Brown]

It gets 1 action and can multitask like a character does. The rules also state that is how it works. Even if you have multiple RDCs, you are multitasking for using each separate one if doing so simultaneously, so that's six of one, half dozen of the other.

If that is what the rules say, then quote the rule; and take your snippy attitude somewhere else. Under RDCs the penalties for multi-tasking are part of the rule of 'a single RDC controlling multiple drones'; and that a robot needs to be running Electronics (Remote Operations) software to use an RDC.

A computer can run several copies of the same software to execute several similar tasks without a multi-tasking penalty; and nowhere is it ever made clear whether or not robot brains are treated differently -- but the ruling assumption is that they operate very similarly to computers. A computer could easily run several copies of E(RO) to drive a matching number of RDCs, each controlling one drone, with no penalties whatsoever -- and nothing about that is controversial. Similarly, with a single (virtual gunner) program a computer could fire dozens of separate turrets at various targets -- and again, there is no multi-tasking penalty. Just to add clarity (/s), robot brains are specifically allowed to run standard ship's computer software. So the question of what exactly the special {Ship's Brain} case of robot brain can do, and what counts as an action, and what situations it gets (and does not get) penalties is very relevant.

 

[Condottiere]

This is probably somewhat more complex than it looks, and probably needs parcelling out.

Speaking of which, sending out packets of information to various parts of the ship, if you don't have the bandwidth, for any particular reason, means lag - that can be perceived as penalties to success.

 

[MasterGwydion]

So basically everyone, including the book's author, has said that you treat it like a character. It gets 1 action and can multitask like a character does. The rules also state that is how it works. Even if you have multiple RDCs, you are multitasking for using each separate one if doing so simultaneously, so that's six of one, half dozen of the other.

But if that isn't good enough for you, then make up your own answer. If you want it to be a fully automated starship, do that.

The book's author can say whatever he wants as to his intent, but RAI is very different in most cases from RAW. We can't play a game based on what is in some author's head and not written down anywhere, especially when this is contradicted by rules or examples of rules in use as written in the books.

So calm your tits and let's get back to being polite.

 

[tytalan]

Robots pg 103
“
Operational Considerations
Any ship’s brain with interfaces can run standard spacecraft software packages if it has available Bandwidth. These packages do not provide additional functionality. For example, to perform
a jump, a ship’s brain running Jump/1 must use Bandwidth 5 to compute the jump and must keep that program in memory while running an Astrogation skill package to complete the jump calculations then, with the Jump/1 program still running, use an Engineer (j-drive) skill package to perform the jump. However, a ship’s brain running Fire Control/1 software can
fire one turret without a Gunner (turret) skill package, as the Fire Control/1 software directly allows for the operation of one turret.”

Also you have to install a robot chassis for the Ships Brain tho a Size 6 or smaller takes no tonnage.

 

[Geir]

Robots pg 103
“
Operational Considerations
Any ship’s brain with interfaces can run standard spacecraft software packages if it has available Bandwidth. These packages do not provide additional functionality. For example, to perform
a jump, a ship’s brain running Jump/1 must use Bandwidth 5 to compute the jump and must keep that program in memory while running an Astrogation skill package to complete the jump calculations then, with the Jump/1 program still running, use an Engineer (j-drive) skill package to perform the jump. However, a ship’s brain running Fire Control/1 software can
fire one turret without a Gunner (turret) skill package, as the Fire Control/1 software directly allows for the operation of one turret.”

Also you have to install a robot chassis for the Ships Brain tho a Size 6 or smaller takes no tonnage.

Wow, so I did say that... I guess it is RAW. I finished writing Robots in June of 2021 and quite a lot has happened since then, so I'm sorry if I can't remember everything - getting to be a theme as I age. Hey, at least I still have my teeth. But as far as the rules go, I think I've said before that that the only time the OTU matters to me is if I'm writing for it.

As far as RAW goes, it's sort of the same thing. Not everybody has to agree with an interpretation, but it is helpful to maintain a consistent interpretation across a campaign and for the Referee and the players to have a consistent understanding of how it works. If you want to do it your own way or the way you think is best, then that's fine. The Traveller police will not come looking for you.

 

[Vormaerin]

The book's author can say whatever he wants as to his intent, but RAI is very different in most cases from RAW. We can't play a game based on what is in some author's head and not written down anywhere, especially when this is contradicted by rules or examples of rules in use as written in the books.

So calm your tits and let's get back to being polite.

I was not intending to be impolite. I'm sorry if it came across that way.

Imho, there's no solution to this issue. He asked how the thing worked. Unless I missed a post, everyone answered consistently that the ship's brain is a single entity and I gave the rules as written about how they work with multiple actions. He doesn't agree and/or thinks there are other ways to do it than that. That's fine. But that isn't written down anywhere how to do that, so you can't get RAW. So he should just do it the way he likes.

 

[J. L. Brown]

That is not what happened. I asked:

So -- does anyone have an answer to 'How many separate actions can a Ship's Brain engage in each turn'? If the brain is in a ship with dozens of turrets, can it fire all of them, without penalty, while also angling deflectors, jamming enemy missiles, and evading fire? Or is it limited to just a set number of simultaneous actions -- and how do we figure out what that number is?

There were several different answers.. Arkathan said 'Virtual Crew rating'; Geir said 'Treat it as a character, so by itself, it is limited to one thing at a time...'. You said:

None of those things require "actions" by the character manning that station or even require anyone to be manning that station. Actions are things like Electronic Warfare or putting the engines in Overdrive. You don't have to have someone doing the "engineer" job to generate Thrust. Firing a Turret weapon does require an "action". Hailing Frequencies Open is not an "action".

And that is obviously incorrect, since 'Open Hailing Frequencies' is explicitly called out as an action in the core rules; as does piloting the ship; and I have yet to meet a GM who would let the players ask "What do the ship's sensors show?" without requiring someone to actually be at the sensor station.

You went on to quote the example brains at the back of the book, and how they operate (via a Robotic Drone Controller), conveniently ignoring that 1} that is a limitation placed specifically on RDC's which are used to operate several drones from the same specific RDC, 2} that the rules for multiple drones through an RDC are different from and contradictory to the rules for a character undertaking multiple actions in a single round, and 3} Avatar controllers -- which may also act as Drone Controllers -- allow multiple Avatars to be controlled simultaneously with no penalties whatsoever. In all cases the actual capabilities of a robotic Ship's Brain (when directly controlling the ship it is built into, through a standard Ship's Brain Interface, with or without the Haptic option) are left ambiguous.

I followed up with:

So the question of what exactly the special {Ship's Brain} case of robot brain can do, and what counts as an action, and what situations it gets (and does not get) penalties is very relevant.

and:

Nor is it clear whether (or how) the TL of the Ship's Brain affects the number of tasks or actions per round.

You said the RAW was clear and simple, and I asked you to provide the actual rule. You have not done so.

Personally I am happy to concede that no RAW solution exists, simply because the rules for this have never actually been written down. I would be very happy if there was a central, Mongoose-run and curated, repository of errata and addenda to cover situations like this.

 

[Vormaerin]

Mongoose has tasks for lots of things that are only expected to be rolled under high stress conditions. There is a task for landing the ship safely at a downport. If you take that literally, you will crash pretty regularly. Further, any non physical task can be done from any workstation.

A scout/courier pilot can certainly operate the sensors, control the engines, use the comms. He's the only guy on the ship. Your GM seriously says "oops, you are flying blind because you can't use the sensors while you are piloting?" Whatever works at your table, but wow.

Anyway, there is no RAW on the issues you bring up and, personally, I think letting the GM decide how they want it to work at their table is a lot better than a cabal of Mongoose employees putting out ever more edge case interpretations. I've never seen a game company actually do anything like that without making the rules worse and more contradictory.