NEWS

Impact of PI-Tacs and Similar Monitoring

  • 29 Replies
  • 13534 Views

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« on: <05-08-14/2243:22> »
Bear with a GM who hasn't run a Shadowrun game since 3rd edition as he tries to figure out this shiny new wireless setting and tech curve.

Even before the new Run & Gun introduced Personal Integrated Tactical Networks,  it seemed to me that most installations are likely to have their guards closely monitored in some fashion.  With commlink cameras, smartlink weapon systems and biomonitors all easily linked and tracked within a network, is there any real means to engage someone without their allies being alerted immediately afterward?

Am I mistaken in thinking the most direct answer would be to have a decker/technomancer brick the device?  Though of course that would still register as someone being offline and thus likely throwing up an immediate flag.  Otherwise (if I'm not misunderstanding the economy of Matrix actions), it would take a Spoof Command action for each monitored device on each guard to fool it into reporting an "A-OK" or freezing/looping records (assuming that functionality is even something the device can do in the first place) before engaging without risk of a guard's PAN detecting any harm to their person, any shots fired or even just catching the Runners on camera.

In a similar vein, other than the PI-Tac flavor text mentioning team networking being subject to malfunctions (something I can't say I've seen a game mechanic for), is there any real reason the team's resident decker/technomancer can't already monitor all the same things as a PI-Tac?  Obviously without the actual mechanical benefits afforded by the PI-Tac system itself, I was operating under the idea that the same functionality was already possible.  Especially if each member of the team has slaved their devices to their commlinks and then slaved their commlinks to the hacker's for protection and monitoring, as my players do.

RHat

  • *
  • Prime Runner
  • *****
  • Posts: 6317
« Reply #1 on: <05-08-14/2304:52> »
Especially if each member of the team has slaved their devices to their commlinks and then slaved their commlinks to the hacker's for protection and monitoring, as my players do.

It's unclear if this daisy-chaining works, but if it does it's actually a pretty bad idea - per the rules, a device slaved to a commlink uses the commlink's attributes to defend; if the commlink is slaved to a more secure system, it's own attributes have not changed.  As such, defending, say, a Smartgun slave to a commlink slaved to a deck is not done with the deck's attributes.  However, if the smartgun is marked, a mark is gained on the commlink, which leads to a mark being gained on the deck.  If it works by the rules, it's actually a pretty insecure setup.
"Speech"
Thoughts
Matrix <<Text>> "Speech"
Spirits and Sprites

SlowDeck

  • *
  • Ace Runner
  • ****
  • Posts: 1015
  • How do people add personal text under their name?
« Reply #2 on: <05-08-14/2309:31> »
I think a big problem with this is that it gives guards a serious upgrade, to the point they could be a serious threat to runners. Which is a problem since the average guard isn't supposed to be.

By fluff, guards are primarily to deal with civilians, not shadowrunners; against shadowrunners, they're basically supposed to be meat shields and cannon fodder to delay the runners until the actual threat shows up. And by fluff, they know it. So something like this isn't actually something I think most guards would have access to.
"Speech" Spirit/"Astral" Thought/"Subvocal" Matrix/"Commlink" "Totem" [Time/Date] <<Text&email>>

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« Reply #3 on: <05-09-14/0134:46> »
Especially if each member of the team has slaved their devices to their commlinks and then slaved their commlinks to the hacker's for protection and monitoring, as my players do.

It's unclear if this daisy-chaining works, but if it does it's actually a pretty bad idea - per the rules, a device slaved to a commlink uses the commlink's attributes to defend; if the commlink is slaved to a more secure system, it's own attributes have not changed.  As such, defending, say, a Smartgun slave to a commlink slaved to a deck is not done with the deck's attributes.  However, if the smartgun is marked, a mark is gained on the commlink, which leads to a mark being gained on the deck.  If it works by the rules, it's actually a pretty insecure setup.

I agree with you entirely on this point, and I certainly don't intend to let the chain-slaving of devices give free reign to exceed the usual limits.  In this case, it was more a matter of monitoring purposes.  For example, if one PC's commlink is monitoring their smartlink weapon system and biomonitor, and the commlink is slaved to the decker's deck, then it appears as though the decker would be able to view those monitors as well.

Mostly it was a statement aimed at how easy it seems to be for a group of people to have at least one character able to monitor a group of others, and thus the ease with which it would seem both Shadowrunners and even the average group of guards have in keeping tabs on one another.  That leads into me agreeing with SlowDeck as well, in that it really ramps up the challenge, but I can't see any reason at all why everyone wouldn't do so, with or without a full-fledged PI-Tac.

SlowDeck

  • *
  • Ace Runner
  • ****
  • Posts: 1015
  • How do people add personal text under their name?
« Reply #4 on: <05-09-14/0140:12> »
Cost. The average guard is, IIRC, armed with less than ¥1000 of equipment. The PI-Tac, when you divide the cost around the group, is worth a few times that. It becomes cost-prohibitive when you're potentially employing thousands or even millions of guards.
"Speech" Spirit/"Astral" Thought/"Subvocal" Matrix/"Commlink" "Totem" [Time/Date] <<Text&email>>

RHat

  • *
  • Prime Runner
  • *****
  • Posts: 6317
« Reply #5 on: <05-09-14/0149:34> »
I agree with you entirely on this point, and I certainly don't intend to let the chain-slaving of devices give free reign to exceed the usual limits.  In this case, it was more a matter of monitoring purposes.  For example, if one PC's commlink is monitoring their smartlink weapon system and biomonitor, and the commlink is slaved to the decker's deck, then it appears as though the decker would be able to view those monitors as well.

Slaving isn't needed to share feeds.
"Speech"
Thoughts
Matrix <<Text>> "Speech"
Spirits and Sprites

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« Reply #6 on: <05-09-14/0154:35> »
Cost. The average guard is, IIRC, armed with less than ¥1000 of equipment. The PI-Tac, when you divide the cost around the group, is worth a few times that. It becomes cost-prohibitive when you're potentially employing thousands or even millions of guards.

Oh, most definitely.  That's why I specified with or without a PI-Tac.  Near as I can tell, you can use a simple Commlink to perform the bulk of a PI-Tac's functionality if you aren't after the actual mechanical benefits of an extra Perception die or such.  There are numerous examples of deckers or riggers keeping tabs on their team long before Run & Gun came out.  I can't find anything at all to prohibit an entire group of guards (or Shadowrunners) sharing access to their various monitoring systems so they can keep tabs on one another, even if all they have is the built-in camera of said Commlink itself.  It certainly leaves them more at risk of a hacking attempt, but it seems as though that risk is worth the complications it leaves your potential opponents.

I can (and am so far) handwave the whole thing and say they don't do so just because it would be a massive pain in any Shadowrunner's butt, but I thought perhaps there was a rule or something I was missing to explain why they don't bother.

Slaving isn't needed to share feeds.

Ah, thank you.  I wasn't sure if it was, and that lends more pondering on my part as to why even the lowest group of grunts wouldn't keep tabs on one another.

RHat

  • *
  • Prime Runner
  • *****
  • Posts: 6317
« Reply #7 on: <05-09-14/0211:25> »
So far as corp guards go, part of what you need to keep in mind is that they'd want to have a minimal cost outlay tied up in these guys, and they wouldn't want anyone who happened to hack one guard's commlink to have eyes all over the facility.
"Speech"
Thoughts
Matrix <<Text>> "Speech"
Spirits and Sprites

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« Reply #8 on: <05-09-14/0230:00> »
So far as corp guards go, part of what you need to keep in mind is that they'd want to have a minimal cost outlay tied up in these guys, and they wouldn't want anyone who happened to hack one guard's commlink to have eyes all over the facility.

That's a good point, thank you.  So clearly you wouldn't want every guard to be running around with streaming details about the rest.  Though I'm curious if that means a PI-Tac wouldn't afford the very same opportunity and provide the same risks, since clearly it would have to stream the information to every member of a team to provide the benefits.

It seems far more likely then, that there would be a person or two serving as a "security room" to monitor the other guards' feed, which means a hacker would have a more concentrated target to pursue.  That comes across as far more doable.

RHat

  • *
  • Prime Runner
  • *****
  • Posts: 6317
« Reply #9 on: <05-09-14/0241:30> »
That's a good point, thank you.  So clearly you wouldn't want every guard to be running around with streaming details about the rest.  Though I'm curious if that means a PI-Tac wouldn't afford the very same opportunity and provide the same risks, since clearly it would have to stream the information to every member of a team to provide the benefits.

It seems far more likely then, that there would be a person or two serving as a "security room" to monitor the other guards' feed, which means a hacker would have a more concentrated target to pursue.  That comes across as far more doable.

Still leaves the security room vulnerable.  Would make more sense just to run a sensor array.

As for the PI-Tac, you probably have to hack the host device, but all the same there's no way corpsec is gonna have that.
"Speech"
Thoughts
Matrix <<Text>> "Speech"
Spirits and Sprites

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« Reply #10 on: <05-09-14/0319:35> »
Still leaves the security room vulnerable.  Would make more sense just to run a sensor array.

As for the PI-Tac, you probably have to hack the host device, but all the same there's no way corpsec is gonna have that.

I see your point about using a sensor array.  It would still be as vulnerable as someone serving as a "security room," but at least it wouldn't provide additional information to a hacker.

This may be topic drift, but as for hacking a PI-Tac, it seems to me you could use Snoop on any device receiving its datastreams, which does seem to make it a rather sizable risk to use if you suspect you'll be facing enemy hackers.

RHat

  • *
  • Prime Runner
  • *****
  • Posts: 6317
« Reply #11 on: <05-09-14/0328:22> »
I see your point about using a sensor array.  It would still be as vulnerable as someone serving as a "security room," but at least it wouldn't provide additional information to a hacker.

This may be topic drift, but as for hacking a PI-Tac, it seems to me you could use Snoop on any device receiving its datastreams, which does seem to make it a rather sizable risk to use if you suspect you'll be facing enemy hackers.

The array would be slaved to the host, so it has those ratings to protect it - the only vulnerability there is the direct connection exploit.

As for the PI-Tac, I'm not sure if the Snoop action is intended to work on it.
"Speech"
Thoughts
Matrix <<Text>> "Speech"
Spirits and Sprites

Khenti

  • *
  • Newb
  • *
  • Posts: 9
« Reply #12 on: <05-09-14/0335:43> »
As for the PI-Tac, I'm not sure if the Snoop action is intended to work on it.

I haven't found anything yet that would suggest it wouldn't.  Though that's probably a question I should put up in the Gear thread.

Michael Chandra

  • *
  • Catalyst Demo Team
  • Prime Runner
  • ***
  • Posts: 9944
  • Question-slicing ninja
« Reply #13 on: <05-09-14/0533:49> »
Tacnet costs around 100k per user. It's why I only am designing my PR5+ KE SWAT teams to have it, not the PR4 teams. The PR5 teams carry a LOT of expensive equipment, worth near-2m new. The PR4 teams not so much, just heavily-modded SWAT Armor and guns.

I can, however, understand secret corps with an important-enough operation that it involves FBA for the Guards, to use a RIG system and have a Spider and some Agents observe the results. But that's PR4+ Security, not ordinary guards.
How am I not part of the forum?? O_O I am both active and angry!

firebug

  • *
  • Errata Coordinator
  • Prime Runner
  • *****
  • Posts: 2996
  • Scraping the bottom of the Resonance Barrel
« Reply #14 on: <05-09-14/1134:24> »
Yeah, allow me to add the the chorus of "That shit's way too expensive for normal corpsec to have."

A level 1 PI-Tac is nearly useless, since the only thing it does that commlinks can't do is a +1 to Perception and a GPS location system.  The rest can be done already.  Any imaging device can share its readout wirelessly already.  Unless Team Biomonitor somehow doesn't require actual biomonitors (implying the PI-Tac adds hardware to everything slaved to it somehow) then that's doable simply by virtue of having biomonitors, which all runners and corpsec should have.  Weapon status indicator, again all weapons do that already.  Combined with it needing to be bought again for every 6 guards on duty, no company would be willing to pay for that crap.

Level 2 has much better benefits; however it is way too expensive for anything but military or HTR in my opinion.

Level 3 makes me laugh.  That stupid "remote and limited access to vehicles and drones".  Guys...  You realize that you can just give other people three marks?  Or just one mark if all you want them to be able to do is "[give] simple commands such as 'go to this location' and 'attack this target'"?  That does not count as extra functionality!  That is like advertising that your DVD has interactive menus as a bonus feature!

As for information sharing, there's no real reason for the corpsec to not be set up to flip an alert as soon as they detect something.  Simple Action, Send Message, "I think I see someone."  If you knock them out before that, their Biomonitor (which may as well be broadcasting to other guards or even the Host, not a huge point in having one if you're not using it wirelessly) will tell everyone they were injured.  It's the information age right now in real life.  This is why a hacker is needed; they can prevent that guard from sending a message or edit the readout of their biomonitor (but that'd only work for a short time).
I'm Madpath Moth on reddit (and other sites).  Feel free to PM me errata questions!
Jeeze.  It would almost sound stupid until you realize we're talking about an immortal elf clown sword fighting a dragon ghost in a mall.