Full Version: How many agents can a decker run simultaneously?
Thread title says it all.
agents aren't tied to anything. they magically steal resources from whatever host they're on (part of the logon process, i guess, except that they don't have to logon to the host they're initially uploaded to... agh! no more thinking! thinking + Matrix = teh bad!)
anyway. basically, you can run as many smart frames and agents as you want. there's no upper limit. you can upload multiple copies of the same frame, too. if you wanted to houserule a limit, i'd say something like (1/2 MPCP), and use terminology stolen from the rigging rules (you can have an unlimited subscriber list, but only half your MPCP in active subscriptions, etc.).
anyway. basically, you can run as many smart frames and agents as you want. there's no upper limit. you can upload multiple copies of the same frame, too. if you wanted to houserule a limit, i'd say something like (1/2 MPCP), and use terminology stolen from the rigging rules (you can have an unlimited subscriber list, but only half your MPCP in active subscriptions, etc.).
With agents as useful as they are, I think a limit of (MPCP/2) is a reasonable house rule.
But then... your deck doesn't need to be online for the agent to operate. It's just a file you upload to the host, which them starts grabbing resources. You can tell it to do anything you want and it'll try and do it even when you're gone.
Anyway, with the house rule you could upload MPCP/2 agents, and tell them all "wait here for 5 minutes, then log on to host SK-711834 from here" or whatever. Then go to that host, log on, upload agents, etc.
It is a problem though.
Bear in mind that uploading a decent agent takes quite a long time, all those null operations could give you away eventually. Or the first agent you upload could give himself away before you get the rest loaded.
Of course, there's at least one way round that too. Aaah.
Anyway, with the house rule you could upload MPCP/2 agents, and tell them all "wait here for 5 minutes, then log on to host SK-711834 from here" or whatever. Then go to that host, log on, upload agents, etc.
It is a problem though.
Bear in mind that uploading a decent agent takes quite a long time, all those null operations could give you away eventually. Or the first agent you upload could give himself away before you get the rest loaded.
Of course, there's at least one way round that too. Aaah.
I'm a GM with a PC who's real fond of agents. He doesn't want to munchkin his character (or their current run) but he can't find anything to stop him from uploading a dozen agents to help with matrix legwork and such.
If you use the actual Matrix Search rules in "Matrix"(pp 129-132) , then Agents really can't do searches by their lonesome, but they can greatly aid Matrix Searches for the decker in question. The limitation is the decker's Intelligence attribute (and sanity) as far as the number of Searches you can simultaneously run.
| QUOTE (Zeel De Mort) |
| Bear in mind that uploading a decent agent takes quite a long time, all those null operations could give you away eventually. Or the first agent you upload could give himself away before you get the rest loaded. Of course, there's at least one way round that too. Aaah. |
If you want multiple copies of the same agent, then just upload it once and start making copies of it on the Host before before you execute it. It would be faster to copy it locally on the Host than to upload it several times.
Also, if you have it compressed and uncompress it on the Host, then you're saving time by uploading a smaller file.
There is more than one way of getting this done in the shortest time possible.
I would have to say that from more of an rp side you can't possible run an unlimited number of agents at the same time you would go crazy very fast but I don't think there should be a limit to how many you can control at one time because there are ingame reasons that stop this like detection and deletion the two d's that prevent anything like an unlimited number of agents from happening either that or a network crash.
| QUOTE (Vertaxis666) | ||
If you want multiple copies of the same agent, then just upload it once and start making copies of it on the Host before before you execute it. It would be faster to copy it locally on the Host than to upload it several times. Also, if you have it compressed and uncompress it on the Host, then you're saving time by uploading a smaller file. There is more than one way of getting this done in the shortest time possible. |
Yep, that's why I said there's at least one way round it.
Your GM could argue that you can't copy the agent when it's already running, but if not that's definitely an option.
Also, if you do have to load them one at a time, putting the Squeeze option on them means they won't run until you spend a complex action to decompress them.
Or just sit on a Blue-4 host and load them all day long there, then send them off to do your bidding.
Also, isn't it kindof handy for Agents that they can have IC options? I wish I could give my own icon the Shift option for example! In a lot of ways, very high level agents are a better bet than SKs. There's no upper limit to their rating like SKs have (other than Computer (Programming) of course), you don't lose the whole program forever if it's crashed (i.e. you still have the source), they don't need a powerful host to run on, etc.
SKs are good, but Utility Pool seems a pretty bad way for them to have to run non-operational utilities. It's very versatile, but would also run out quickly if you had to make a few different tests in one phase.
| QUOTE (mfb) |
| agh! no more thinking! thinking + Matrix = teh bad! |
only if you mix real life and matrix computer knowledge in the same train of thought
So true!
I did my degree in Computing Science, yet I'm quite at ease with the Matrix rules. Just don't think about it all too much and you'll be fine.
I did my degree in Computing Science, yet I'm quite at ease with the Matrix rules. Just don't think about it all too much and you'll be fine.
Is it just me, or does this thread conjure up images of DoS attacks?
pretty much what i'm imagining would happen too Nick, although, i dont know if there's any rules in matrix for a host recognising an abnormal amount of traffic
As far as I know there are no rules for that.
Although with the hundred agent scenario, it's not going to take long before one of them triggers a Passive Alert, which will then affect the whole system. In time it could go to Active Alert, or even shut down totally, with that many Agents running about.
Although with the hundred agent scenario, it's not going to take long before one of them triggers a Passive Alert, which will then affect the whole system. In time it could go to Active Alert, or even shut down totally, with that many Agents running about.
| QUOTE (Aku) |
| [...]i dont know if there's any rules in matrix for a host recognising an abnormal amount of traffic |
Actually this is done during every test the agents have to make. That's one of the reasons why they will build up a security tally....
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.