[SR4] Matrix Clarification Help

[AnimeTheCat]

Hello all. I originally posted the below to reddit and... yeah... it was less than helpful. I've had a better experience on this forum for nearly everything I've tried, so hopefully I thought it wouldn't hurt to post here too. Link to original Reddit post. As with the original post, if you know of a better place to ask this, let me know.

Hello all. Brand new to this sub, but not to shadowrun. I'm like, reborn new, not brand new. The last time I played was 9 years ago when my brother was the GM. I didn't care for the Astral or the Matrix so I played a gunbunny adept and shot things for craptons of damage from really long ranges. Was fun. Now I'm the GM and I have to wrap my head around everything, which is easy enough now 9 years later. I read through the awakened world in the SR4A book and got it immediately. Was simple enough. Meat space was always easy. The matrix isn't really all that difficult either, I'm only having a hangup when it comes to nodes, personas, accounts, programs, and agents. So let me explain it how I think it goes and any matrix wizes out there can tell me how wrong I've got it.

So, you've got nodes. Nodes are everything from your toothpaste tube the satelites in low earth orbit. If it's not a node, it's an icon. Nodes have 4 "attributes", system, firewall, response, and signal. Of them, I find system to be the most important - in general. System restricts the number of programs that can be running on the node via the Processor limit (system for normal nodes, system x2 for nexus), as well as the maximum rating of those programs (a system 4 node can only run a Search 6 program at rating 4, limiting the dice pool bonus to +4 from the program). Additionally, personas inherit their rating from the node they're present in.

Personas are the firmware that represents you in the matrix. It inherits it's system rating from the node it's installed on. Personas project and Icon of you on to the matrix which you use to interact with other icons and nodes. Personas can be loaded with as many programs as desired by the character. When a persona decides to change a program from loaded to running, they spend a complex action to start the program, at which point it begins running on their node. If you have an account logged in to another node, you can run programs on that node depending on your access permissions. When run in this way, you run a program loaded in your persona on the node you've accessed, thus using up that node's processing limit rather than the node your persona is in. Your persona also has a limited number of subscriptions it can maintain equal to twice your persona's system rating.

Agents are semi-autonomous programs that can interact with other programs and icons. They are described as "Drones of the matrix" because they can be controlled similarly to drones in meatspace. As they are programs, they count against the Processor limit of whatever node they are running in. An agent can be running in the same node as your persona and are able to operate functions for you using your access ID and account. When used this way, they have the accesses you do and function at the same level as your persona's node. Agents running in the node your Persona is in count against that node's Processor limit, but not your Persona's Subscription Limit. Agents can also be running in different nodes than the one you are in, following specific commands as directed. Giving a single command to any number of agents is a simple action. Different commands require different simple actions. To operate in a different node than you are in, you must either run the agent in your node and command it to go to another node, or you must run it in that other node, which requires you to have the appropriate account. Agents operating independently from the node which your persona resides in count against your persona's subscription limit, but not your node's processor limit. Programs that agents run count against the processor limit of the node they are running in.

For example, you have a run that is to bust a mobster out of a low security prison block. 3 guards, one security node, one medical drone, minimal resistance expected. The group decides it's best to get in and get out with as little shots fired as possible, they don't want Lone Star on your tails.

let's say you're hacking with a stock Hermes Ikon running the Novatech Navi OS. Your Hardware gives your commlink response 4/signal 3 and your OS gives it Firewall 3/System 4. Your persona has all common use and hacker programs loaded at rating 4, as well as 3 rating 4 agents each with full copies of your software library. You know you're going to be hacking on the fly, so you prepare by going in hot sim to give you better matrix initiative and +2 matrix IPs. You always have an agent running on your node that is always running analyze and scan, the agent is designated to analyze your node every turn and alert you to intruders.

Spoiler: Turn by Turn Actions

Show

1. Turn 1, IP 1 - You start by detecting hidden nodes (complex action). Since you know where you're looking, you only have to succeed a threshold 4 test, which you do. Your agent analyzes your node and find nothing unusual.
2. Turn 1, IP 2 - You stop running Scan (simple action) and command your agent (simple action) to run the exploit program and designate it for your use. The agent runs the program (complex action).
3. Turn 1, IP 3 - You analyze the node (simple action) you discovered and command your agent (simple action) to run the Stealth program. Your analyze returns 3 hits, which you use to ask if the node is encrypted, what the firewall rating is, and what the system rating is. The node is encrypted and the the firewall and system are both rating 4. The agent runs the stealth program (complex action), which brings you to 4 running programs, so your response is reduced to 3.
4. Turn 2, IP 1 - You issue two commands to your agent (2 simple actions), first is to begin running the decrypt program, then after that to run the edit program. Your agent runs the decrypt program (complex action), you're now running 5 programs.
5. Turn 2, IP 2 - You begin hacking on the fly (complex action) to create a Security Account on the node. You use a point of edge before you make the check because time is of the essence, and you manage to create your illegitimate security account on the first try, nice. Also, it doesn't appear that the firewall detected you over your stealth. Your agent runs the Edit program (complex action). You have 6 programs running on your node now.
6. Turn 2, IP 3 - You Log In (complex action) to the node you just created the security account for. Your agent analyzes your home node and doesn't report anything.
7. Turn 3, IP 1 - You command your agent (simple action) to initiate cryptanalysis on the node and you stop running (simple action) your exploit program. Your agent initiates crytanalysis (complex action). You're down to 5 programs on your node.
8. Turn 3, IP 2 - You create a legitimate security account (Complex Action). Your agent analyzes your home node and reports nothing. The cryptanalysis completes.
9. Turn 3, IP 3 - You give your agent 2 commands (2 simple actions) - log in to the legitimate security account and create another security account for you. The agent leaves your node bringing your programs down to 4 on your node.
10. Turn 4, IP 1 - You analyze the node (simple action) and find a patrolling spider as well as exit points to the 3 man squad's commlinks, the security cameras, and the prison doors. You stop running edit (simple action) bringing you down to 3 programs and back up to 4 response. Your agent runs edit (complex action) on the node it's just moved to.
11. Turn 4, IP 2 - You stop running decrypt (simple action) and command your agent (simple action) to start an agent program on the node it's on once it has completed the account creation. You're now at 2 programs on your node (analyze and stealth). The agent successfully rolls pilot+edit (3) to create a security account.
12. Turn 4, IP 3 - You run an additional Agent (complex action) in the newly created security account. Your agent runs a third agent on the node. At this point The spider analyzes the node and notices your stealthed icon and alerts the 3 man guard team that there's matrix trouble.
13. Turn 5, IP 1 - You order the first agent to work with it's agent to take out the spider (simple action) and the second agent to begin running defensive programs, starting with armor, and designate them for your use. Agent 1 runs the attack program (complex action), agent 2 Attacks the spider making 1 net hit, and so deals 5 damage, 1 of which is resisted (the spider is not running armor yet), and agent 3 runs armor and designates it for your use. The spider panics and begins attempting to reboot the node.
14. Turn 5, IP 2 - You send a command to the commlinks of the guards to eject all magazines, and a second command to the goggles to stop the image link processing (2 simple actions, no spoof necessary as the commlinks are slaved to the security node and accept commands from security accounts). Agent 1 attacks (complex action) and hits with 2 net hits for a total of 6 damage (2 resisted), Agent 2 attacks (complex action) and hits with 3 net hits for 7 total damage (none resisted). This is sufficient to crash the spider's icon, stopping the reboot cycle. Agent 3 runs the biofeedback filter program (complex action).
15. Turn 5, IP 2 - You command all of your agents to bring as many security systems from connected nodes offline, in no particular order, except for the (simple action) and you order the Evo Orderly drone to make it's way to the cell where the prisoner is being held. Your agents begin sending deactivation and shut down commands to everything on the network (6 commands at a time). (6 simple actions).
16. Turn 5, IP 3 - You jump in to the Evo Orderly and continue to move to the prisoner. Your agents continue to deactivate programs and security features on the network. The chaos makes it easy for the group to move in and subdue the enemies with tazers, gel rounds, and other stun damage.

Before you Logging out of the node, you make sure to edit all of the logs to remove your ID, the ID of your agents, and the accounts. You finish the mission out by modifying your ID so that any missing data can't be traced back to you anymore.

I know this can be made more simple with scripting, which I will help my player(s) develop, but this is sufficient to highlight how I think programs, personas, icons, and agents interact/intersect. Please let me know how on the mark or off the mark I am with this as I think I've managed to figure it out pretty well. Before I go telling my players anything I want second opinions.

Thanks in advance for the help. If there are any better communities to turn to for questions about 4e I would appreciate knowing them as well. I'm not willing to buy any new books (especially considering the state of the world) and I already have SR4A, RC, UW, AG, AR, and SM so that's like... $200 at least to replace for a new edition.

[AugustNights]

If you can pare down that wall of text to a question, or a bullet point of clarifications I'd be happy to offer insight.
You might find folks with less ADHD and Dyslexia than me over at the Dumpshock forums, though.

[AnimeTheCat]

If you can pare down that wall of text to a question, or a bullet point of clarifications I'd be happy to offer insight.
You might find folks with less ADHD and Dyslexia than me over at the Dumpshock forums, though.

Sure, I can do that. I think I've figured out some of it for myself but I'll try to pare it down to some bullet questions.

1. Where do programs run; on personas/agents or on nodes? I arrived at the conclusion that they run on nodes and they have to be running on a node for a persona or agent to use.

2. When you are in a node, does your simple action observe in detail entitle you to a matrix perception test to see icons running stealth, or do you need to use an analyze node/icon simple action?

2.a. if it requires an analyze node/icon action, does this oppose all icons stealth+hacking/firewall or do you have to declare a certain one?

2.b. combining 1 and 2, when you run stealth is it running on the node you're going to or on the node hosting your persona? I.e. the node you're hacking or your commlink?

For now, I think those are my gray areas.

[AugustNights]

1. Yeah, programs run in a node, and are used by a persona or agent.

2. AFB, but I seem to recall that observe is not sufficient to see folks with stealth.

2a. Still AFB, but I don't know how you would single out an Icon you don't know is there. You'd probably roll against all.

2b. My understanding is that you are running the program on your node, not the target node.

[AnimeTheCat]

2b. My understanding is that you are running the program on your node, not the target node.

That's what I thought about the frist three, but now I have a correlary question about this one. If you have an appropriate account to run a certain program (i.e. users can run browse, edit, etc; a Security user can run track, ECCM, etc in addition to general use programs, and an admin can run all programs) on a node, can you run the program on that node even if you're persona is in your commlink.

For example, say I'm running Stealth, Decrypt, and Exploit on my commlink and I successfully decrypt and hack in to a node without being detected. I took the +6 threshold to get an admin account. Can I choose to run an Agent on that node, or will the agent run on my commlink?

[AugustNights]

My understanding is you gain the permission to do action to the node with your own programs, but the programs are always running on your own node.
Cannot confirm at this time, will defo bust out the books soon.

[pragma]

It's been a hot minute since I tried SR4, but I just dug up the core book to take a crack at this. Answering the same questions slightly differently from August:

1. Yes, programs run on nodes. Specific to your question, programs that you and your agents use run on your commlink's node.

2. Finding hidden nodes requires the Detect Hidden Node action (SR4A pp230). Observe in Detail gets you a list of all non-hidden nodes. You can use data search to filter down from a long list of nodes/wireless traffic, but the rules are under-specified: one classic failure point of the system is carrying around, e.g.: a bunch of hidden RFID tags so that someone scanning for your deck needs to guess which of the thousands of nodes that show up on traffic is your commlink.

2a. Analyze Node is always for one node. You can think of it as a rough matrix analog to assensing, down to the table of information you can buy with hits (SR4A pp228)

2b. Commlink node

Followup: You could run whatever program you wanted on a node to which you had admin access. Running an agent on a node you'd hacked would allow you or any other admin to command the agent. That said, I think you can't have two copies of the same program running at once, and Unwired has advanced rules for scaling the idea of running programs on other people's stuff up to Botnet levels.

[AnimeTheCat]

My understanding is you gain the permission to do action to the node with your own programs, but the programs are always running on your own node.
Cannot confirm at this time, will defo bust out the books soon.

Followup: You could run whatever program you wanted on a node to which you had admin access. Running an agent on a node you'd hacked would allow you or any other admin to command the agent. That said, I think you can't have two copies of the same program running at once, and Unwired has advanced rules for scaling the idea of running programs on other people's stuff up to Botnet levels.

Thanks both of you for the help, but these two points seem to be at odds with each other. I'll look and see what I can glean from unwired again, but do you have a section to maybe trim down what I should search through Pragma? And if you were able to find any kind of reference for what you said too, AugustNights, I would appreciate it.

thanks again for the help you two. Though your answers have been the fewest, they've been the most helpful.

[pragma]

I based my reply to your add-on question on the Run Program action on SR4A pp 231: "You run a program, agent or IC that is loaded in a device to which you have access and appropriate privileges."

Unwired pp94 and pp99-101 discuss botnets.