Drunken Review: Shadowrun 5

[Heath Robinson]

Section.11: Street Gear

It's a shame you didn't mention that Free Action Drop Out is a rule in this section:

TURNING IT OFF
Toggling an individual device’s wireless functionality off is a Free Action, as is toggling all of your wireless devices to “wireless off.”

It's a drop in the bucket so far as reasons why the hacker doesn't having hacking actions worth using in combat, but I think it's a nice demonstration of how the book as written directly undermines itself. The left hand really doesn't know what the right is doing, has done, and will do at all.

[Username17]

Yeah, the sheer depth of how much the Matrix rules don't work is really impressive in a way. It's not salvageable, because there are actually a whole lot of nested reasons it doesn't work. The thing where characters can actually cut out their matrix connection at any point as a free action and force enemy hackers to go sit at the children's table is actually a pretty small portion of how messed up this all is. I mean no doubt it really does completely fuck the Hacker, but there are a lot of nested reasons he can't do anything.

The iconography thing is actually the biggest problem, because it's first. The book just sort of assumes that you're sending all your analysis rolls against the icons of important devices, but you literally don't have the information of whether an icon is important or not until after you start analyzing it and succeed at an opposed roll. It's a fundamental cause and effect problem.

Edit: And making this thread reminded me of This Thread from four years ago. Looking at it just makes me sad now.

-Username17

[Neurosis]

TL;DR: you roll magic+summoning, spirit rolls Force*2. For you spirit's Force acts as a Limit. So, when summoning Force 1 spirit you can't get more than 1 hit, meaning you can't get more than 1 service from him.

I believe this chart might be helpful for those who don't like calculating probabilities in their heads: (Spoiler for width)

This chart is in error (or at the very least has an error). The probability of getting 5 hits on 5 dice is NOT 0%. It may be very low but it's not 0. This can happen. I've done it dozens of times myself (but then again my dice luck is ridiculous). Anyway I think we would all agree it's possible.

[Whipstitch]

It's pretty obviously just truncation.

[Blicero]

Yeah, the chart only gives two digits of accuracy. The probability of getting 5 hits on 5 dice is less than 1%. Thus, it's printed as 0%.

[fectin]

The chart is obviously to the nearest %. Anything less than 1 in 200 will show up as 0.

[TarkisFlux]

That's rounding. There is a 0.41% chance that you get 5 hits on 5 dice, which is appropriately rounded to 0% for the display. Chart might be better if it showed to the tenths place, but it would also be a lot more crowded.

[Whipstitch]

The chart's fine as it is, really. In SR you're typically rolling against long odds because you were being forced to and/or because getting a few hits on a test is often still better than abject failure.

[Neurosis]

A second after I posted, I figured it was truncation, but I do wish the chart itself noted that. *shrug* nbd

Would it work if the matrix was abstracted to being "function-oriented" ? I mean:

The group must infiltrate a research complex. [decker rolls for accessing the system and faking personal Ids for everybody]. The group uses the Ids to enter the complex and come across an elevator to an off-limits floor. [decker rolls for activating it]. The group comes across a big hallway full of cameras [decker rolls for disabling it]. The group comes across a safe [decker rolls for opening it]. The group gets the package and get out through a side-door [decker rolls for deleting/manipulating sensors registry to the group presence in the building]. End run.

If each of those decker interventions consist of only one quick roll, and any Ice / intrusion countermeasures / security deckers would just be modifiers to that roll, it would make the whole process super-fast and also make it possible for the decker to be off-site (since he would be interacting with the group through normal conversation as if he was present anyway).

Thoughts ?

This bullshit is basically what a lot of people did with Shadowrun. "Just roll Computers", ignoring the decking rules entirely. It's also basically how I run Hero System (the skill system is I pick a skill, you roll that skill if you got it, maybe at a penalty depending on how hard what you're trying to do is, if you succeed you succeed) and how I've seen hacking in Eclipse Phase run the one time I played it.

One (and there are more than one) reason that this doesn't really work for Shadowrun is that Intrusion Countermeasures aren't just obstacles/complications, they're also CONSEQUENCES.

[Ferret]

Wait, why is the fact that IC is a consequence a dealbreaker? IC sets difficulty, failure degrees give results based upon the IC in the system. It's a bit more work on the DM's part, but not any worse than D&D Thief's find/remove traps fallouts.

[John Magnum]

Even if it was only worse in degree, that would still be terrible because trapfinding in D&D is a miserable joyless exercise.

[Username17]

Wait, why is the fact that IC is a consequence a dealbreaker? IC sets difficulty, failure degrees give results based upon the IC in the system. It's a bit more work on the DM's part, but not any worse than D&D Thief's find/remove traps fallouts.

It's a deal breaker for silva's suggested system, not for hacking systems in general or even for all (or most) vastly simplified hacking systems. Security measures are staged - setting off alarms causes additional resources to get committed. So the IC can't just be a modifier to your single roll, because some of them aren't active until the results of your roll have been calculated. Major security systems necessarily may require two or three rolls to resolve.

-Username17

[Seerow]

Even if it was only worse in degree, that would still be terrible because trapfinding in D&D is a miserable joyless exercise.

It is, but at least it resolves quickly. Much better having the rogue roll trapfinding once and either succeed or get smacked with the trap than have the rogue go off and do his own whole goddamn dungeon while the rest of the party sits around waiting, just so they can get back to doing their own dungeon once the rogue finishes to open their door.

[Ferret]

It's a deal breaker for silva's suggested system, not for hacking systems in general or even for all (or most) vastly simplified hacking systems.
-Username17

Ah, I see where you're coming from.

In my head, I'm equating <hacker rolls for x> as allowing for more than one roll in series if you wanted. Maybe a Recon (Identify IC/ "Find Traps"), Disable (neutralize IC/"Remove Traps"), Manipulate (make change/copy files/what have you/ "Pick Locks") action chain.

Failures could reference a chart of consequences ranging from System ALERT levels to black ICE dispatch.

Maybe add a couple of generic actions the Hacker can take to lower system warning levels, or maybe have that sort of positive background effect happen on degrees of success "You hack that ICE so hard you backtrack it into the system and reset Alert level to 0" or something. In any event, I'd try to limit Hackerspace to 3ish rolls at a time, so they could be an acceptable speedbump/interlude instead of everybody waiting around while the hacker roots the system for an hour and a half.

Personally, I'd do it in one step as Silva proposed, and a failure would just increase the difficulty of subsequent actions and raise security alarms. A bad failure would do that and damage the hacker or his gear, and critical failures would roll rapid response teams or whatever.

[Hadanelith]

Question: is there a concise, cogent explanation somewhere of how this proportional damage system thingy works? Frank talks about it like everyone knows what it is, and I only sorta get it.

[Lokathor]

Question: is there a concise, cogent explanation somewhere of how this proportional damage system thingy works? Frank talks about it like everyone knows what it is, and I only sorta get it.

http://tgdmb.com/viewtopic.php?p=206244#206244

[Cyberzombie]

In my head, I'm equating <hacker rolls for x> as allowing for more than one roll in series if you wanted. Maybe a Recon (Identify IC/ "Find Traps"), Disable (neutralize IC/"Remove Traps"), Manipulate (make change/copy files/what have you/ "Pick Locks") action chain.

All those shouldn't be rolls. It should just be one roll to do something with a failure meaning that you triggered an IC (assuming one exists). IC can just be zap traps, where the IC nukes you with something and then fades away.

Matrix perception tests should die a horrible death.

[Username17]

You're going to need something called a Matrix Perception Test, because hidden security systems and hackers trying to avoid detection by security spiders are so central to the conceits of the genre. But you need to not have anything that works remotely like any Matrix Perception Test that any edition of Shadowrun has ever thrown around.

Rolling "per file" is simply not acceptable considering that there is by definition a virtually unlimited amount of available files.

-Username17

[TheFlatline]

You're going to need something called a Matrix Perception Test, because hidden security systems and hackers trying to avoid detection by security spiders are so central to the conceits of the genre. But you need to not have anything that works remotely like any Matrix Perception Test that any edition of Shadowrun has ever thrown around.

Rolling "per file" is simply not acceptable considering that there is by definition a virtually unlimited amount of available files.

-Username17

I don't get why things can't have a... shit I don't know stealth rating, either generated by hardware and software or by flat out rating, and your matrix perception test just kind of is a general ping of the area/node/whatever and if you beat the rating you see the... whatever. If you don't meet or beat the score you don't.

Just... sort of like how perception tests work for the rest of RPGs.

[Ancient History]

Even in "the rest of RPGs" it can get problematic, because usually if someone or something is deliberately hiding they want an opposed test - and sometimes one for each person or thing that is hiding!

[TheFlatline]

Even in "the rest of RPGs" it can get problematic, because usually if someone or something is deliberately hiding they want an opposed test - and sometimes one for each person or thing that is hiding!

But isn't that what you normally have for perception anyway?

3 people want to hide. They roll stealth. Boom. 5 hits, 3 hits, 6 hits. Or whatever. Someone wants to peek around and see if someone is hiding. He rolls 4 hits. Sees the guy with 3 hits, doesn't see the guys with 5 or 6.

With the matrix, outside of combat, given enough time, you'll eventually pick up on obfuscated icons. If time is no factor at all I wouldn't even force a roll. But that should be sort of rare- time usually is a factor. So you maybe have your passive analysis- your hardware, maybe your software, generates automatic hits that basically burn through obfuscation hits. So your deck/commlink/whatever gives you 2 auto-hits and maybe your Analyze software gives you another auto-hit. So anything that tries to obfuscate needs to roll (or rates) better than 3 hits to confound your passive systems. Active analysis lets you roll on top of those hits. But it's like active sonar- it sends out a broadcast/blast ping and you get more information back. In real world terms this is basically a port scan, saying "is port 1 through 65 thousand open?" and listening to which ports reply and deciding what functionality is based on the ports that are open and the reply that is received from a query of the port.

Much like a port scan though, active analysis is noisy. It sets off all kinds of alarms, or at least generates attention really, really fast.

I dunno, I'm spending way too much time reading on WW2 submarine warfare and I'm starting to think about the Matrix as a kind of cat & mouse game the way sub warfare & evasion was back in the day.

Edit: I see what you're saying. Some people want a separate perception test for each person in the room. That's kind of a chicken & egg thing though and I never really liked the idea so it wasn't apparent to me immediately.

[Username17]

Even one perception test per sentry and one stealth test per sneaker can get pretty absurd pretty fast. You got the passive system defenses, a security spider, and the IC, and that's three sentries who are likely all also hidden. Then you got any number of hackers and/or their drones and agents.


So right away you have to solve your Agent Smith problems. If the IC is modeled as multiple actors who roll separate perception and stealth checks, you're fucked. But even if it isn't, you could still be looking at a very large pile of die rolls if there are a lot of characters on each side. Of course if you roll perception separately against each target you're looking at quadratic growth of die rolls as the number of actors increase, so that idea is DOA. But still, we're talking about a minimum of 2 die rolls added for each new actor in play. And for a team game with maybe six players who also have henchmen and drones, that could obviously get large even with rather extreme systems in place to prevent proliferation of die rolling.

-Username17

[Cyberzombie]

You're going to need something called a Matrix Perception Test, because hidden security systems and hackers trying to avoid detection by security spiders are so central to the conceits of the genre. But you need to not have anything that works remotely like any Matrix Perception Test that any edition of Shadowrun has ever thrown around.

Rolling "per file" is simply not acceptable considering that there is by definition a virtually unlimited amount of available files.

The only time I can really see having a matrix perception test is when the spider is actively looking for a hacker that he thinks is there (like if one of his security goons says his tech is acting funny), but that the system has yet to detect.

Aside from that one instance, you really don't want to be tossing out perception checks to handle routine patrols. I would just handle most hacking actions as a simple test, with spiders/IC just adding to the threshold of that test, or giving dice pool penalties. Only when you fail at a hacking test do you actually get detected, in which case no perception test would be required since the system has flagged you as causing an alert.

[Username17]

That is completely unworkable. The player characters can be wearing the white hat or the black hat in any particular confrontation. The PCs also get ambushed sometimes, so it is totally unacceptable for any discussion of whether an attack goes undetected to be resolved entirely by rolls by one party or the other.

-Username17

[Cyberzombie]

That is completely unworkable. The player characters can be wearing the white hat or the black hat in any particular confrontation. The PCs also get ambushed sometimes, so it is totally unacceptable for any discussion of whether an attack goes undetected to be resolved entirely by rolls by one party or the other.

Why? That's how computer intrusions generally work. The whole point of many good hacks is pulling them off without being detected. I'd say it's ideal for the game system if the hack target doesn't have to make a perception check, because in addition to reducing die rolls, it also reduces the amount of potential metagaming from PCs when they make perception checks and detect nothing until it's too late.

That's how normal stealth should work too. Your spidey sense doesn't tingle when you fail a perception check, you just have no idea what's going on until the shit hits the fan. So why have that check at all? Just let the guy doing the sneaking make the roll and that's that.