[Adder]

Hey all,

I'm having a lot of trouble understanding the concepts behind the matrix. I'm mostly familiar with previous versions of the (wired) matrix though I won't lie and pretend a lot of my understanding isn't based off of the old Genesis game.

Here are some of the questions I have so far:

1. What is a grid? No, really? Based on the rulebook's explanation I imagine it to be similar to a cellular network. For example, there are towers scattered around the city (e.g. Seattle) that are constantly transmitting/receiving information. Great, that makes sense. If there are two grids (for example Seattle Local Grid and SomeLocalCompany Private Grid) they may both be available in some areas of the city, or maybe just one (for example, a part of the city that has none of that company's offices), or neither (on the city outskirts, deep inside a large building, underground..). If I am in a location that only has access to one grid (e.g. the public grid), can I jump to another grid? If I use the Seattle Grid to connect all the way around the world to the London public grid and the Seattle grid goes down (terrorists, malfunction, whatever) do I lose connection to the London grid? I hope it makes sense why the actual implementation of the grid matters.

2. What is a global grid? Similar to question #1 above, I'm presuming that a global grid is a network of ridiculously large, powerful transmitters/receivers placed around the world.

3. What is a host? The rulebook states that there is no physical presence for a host, and that it's made of the Matrix itself. Does that mean there is no physical object in the real world that "defines" the host? I mean, something has to store what the host "is" and what is "in" it. That could either be the Matrix grid itself, or a machine that connects to the grid (analogous to a RL web server).

4. If a small company had a single computer that they wanted only accessible within the confines of their building, how would that be implemented?


Thanks for reading! I hope my questions don't seem too trivial or silly.

[Namikaze]

Hey all,

I'm having a lot of trouble understanding the concepts behind the matrix. I'm mostly familiar with previous versions of the (wired) matrix though I won't lie and pretend a lot of my understanding isn't based off of the old Genesis game.

Honestly, a lot of people have the same issue.  I'm still trying to wrap my head around it completely, because there seem to be times when I think I've got everything put into these neat little boxes and then someone comes around with an idea that I hadn't prepared my neat little boxes for and things go tits up.  I'll do what I can to explain what I can, but there may be contradictions.  I think the only person on these forums that has any solid understanding of the utility and limits of the Matrix is Xenon, but I haven't seen him post for a while.

1. What is a grid? No, really?

Grids are a throwback to the older rules (anything prior to 4th edition).  Basically, you have to imagine that there is a massive telecommunications network, called a grid, that handles all telecommunications traffic in an area.  There are global grids, which are owned by the Big 10, local grids usually limited to a region like Seattle, and public grids which are practically ad-hoc mesh networks (aka: terribad).  All these grids mesh with each other and work with each other, but if you have to hop a grid there is latency.

This is similar to the current Internet in that there are network backbones, all connected to each other.  If you want to go to Google.com, for instance, you end up bouncing off your ISP and a few other places before you actually get to your destination.  Note that when I say "destination" I mean your goal server.  All data that you send and receive to and from that server has to travel along all those hops.

So you've got a commcode, which is the Shadowrun equivalency of an IP address.  This is what ensures that your data is routed to your device correctly.  You connect via your grid to the Matrix as a whole (which is the combination of all grids and Hosts).  If you're in Seattle and you subscribe to the local grid (medium lifestyle or above) then you have no need for extra hops unless you try to access something that is located on a different grid.  Let's say you want to access Seattle's official Host.  The Host would most likely be located on the local grid, and therefore you have no noticeable latency or problems.  After you get done paying for your parking ticket, you decide to find a local Ares gun store in order to buy a gun to shoot the idiot that parked in your spot.  You are still on the local grid, but you find the Ares host and log in.  Since the Ares host is on the Ares grid, you might receive a little latency while making the connection.

Hosts are whole new thing though - they're like private sub-grids.  So once you're connected to the Host, you have no more Noise while you're in the Host.

2. What is a global grid? Similar to question #1 above, I'm presuming that a global grid is a network of ridiculously large, powerful transmitters/receivers placed around the world.

I think I covered that in my first answer - if you have additional questions though, I'd be happy to tackle them.

3. What is a host? The rulebook states that there is no physical presence for a host, and that it's made of the Matrix itself. Does that mean there is no physical object in the real world that "defines" the host? I mean, something has to store what the host "is" and what is "in" it. That could either be the Matrix grid itself, or a machine that connects to the grid (analogous to a RL web server).

Good question.  We don't know what a Host is physically composed of, if anything.  Basically, a Host is a sub-grid.  You might visit the Seattle government's Host as in the above example, which is a sub-grid of the local Seattle grid.  And then you might go to the Ares Host, which is located on the Ares grid.

4. If a small company had a single computer that they wanted only accessible within the confines of their building, how would that be implemented?

Not at all currently.  Sort of.  We don't have rules for how to develop or manage Hosts yet.  That will be out with Data Trails, I would imagine.  For now, just assume that someone like a mom-and-pop-shop could go and rent Host space from the grid that they're subscribed to.  That is pretty much how I have been treating it.

Thanks for reading! I hope my questions don't seem too trivial or silly.

These are the questions we all have, honestly.  I spent some time trying to adapt what I could for technomancers from 4th edition's Unwired, but I couldn't even begin to try to adapt anything else.  The technomancer stuff breaks the rules of the Matrix by default, so I took some liberties with that.  Because the biggest change to 5th edition is, in my opinion, the changes related to the Matrix, I've been Jonesing for the Data Trails book since day one.  So at least you know you're not alone in this. 

[DeathStrobe]

What I think Grids are are a priority system for data. This is why communicating to devices right next to each other but on different grids suffer a -2 to Matrix actions, because the devices are only communicating with each other when they have idle time. This way the entire Matrix is still running on a mesh network like in SR4, but paying for priority Grid access gets your data a higher priority. Like the public grid has its data packets prioritized the lowest, which is even why it has a -2 even when communicating between 2 devices on the public grid. But who knows if that's how it really works. Hopefully Datatrails will clarify.

As for Hosts, they're build on cloud computing. So they're mass distributed processing existing on hundreds of dozens of devices. This way it'd pseudo still work with my theory of what Grids are. But its also possible that Hosts are build on the very backbone of whatever runs the Matrix, if my assumption is wrong.

[8-bit]

What I think Grids are are a priority system for data. This is why communicating to devices right next to each other but on different grids suffer a -2 to Matrix actions, because the devices are only communicating with each other when they have idle time. This way the entire Matrix is still running on a mesh network like in SR4, but paying for priority Grid access gets your data a higher priority. Like the public grid has its data packets prioritized the lowest, which is even why it has a -2 even when communicating between 2 devices on the public grid. But who knows if that's how it really works. Hopefully Datatrails will clarify.

That's sort of how I envisioned it as well. I think of it as a 3G vs. 4G and different providers.

The public grid has access only to the lower bandwidth and low priority equivalent of 3G, which is why it's always so slow and laggy. Whereas talking to other devices across grids is like an AT&T Phone calling a Verizon Phone (although vastly more complicated, it's not just a call). The two providers have to talk to each other and provide a key to allow access to each other, as well as coordinate data priority levels, thus giving some lag and a penalty.

[Sabato Kuroi]

In Splintered State, Brackhaven Investments HQ have 2 hosts.One is accessible wirelessly (security host) and one is accessible only with a direct connection (data host).So I Imagine a small company would do the same thing.A small server room with a private host.

Not an expert on the Matrix or  rl computers though :/

[Kincaid]

4. If a small company had a single computer that they wanted only accessible within the confines of their building, how would that be implemented?

There aren't currently rules for it, but the core book mentions, "executives have learned to store some of their private data in wired nodes, meaning the runners have to track down those specific pieces of machinery..." (SR5, p 53) as well as "hardwired security" (ibid, p. 15), so these things certainly exist.

[Adder]

So to be clear, if the grid you are currently on goes down, you would also lose connection, yes? As would any intermediary grid between you and the destination grid?

Also, if I connected to a fantastic grid from a lousy one, shouldn't my traffic still be capped to the lowest common denominator (i.e. the lousy grid)?

It's hard for me to design the matrix component of runs as a GM. Even something as simple as the runners breaking into a small office and collecting data from a computer leads to all sorts of odd questions. Is the computer also a host? Is it accessible from the grid? Is it only accessible when you're close to it? Wouldn't that require it's own transmitter/receiver (basically, wi-fi) and it would kind of be its own grid? I'm not sure if that makes any sense, but it's actually the really mundane nuts and bolts I'm having trouble with.

I'm actually a web/software developer for a living so I tend to see all things from that perspective.

Let's imagine another example, like a bullet train traveling through a remote area. The train can be controlled from the Matrix, and passengers can also use the train's "wi-fi" to connect to more distant grids. How would this be implemented?

Is the train a host? Is the train a grid? Is there a computer on this train that I could destroy to "destroy" the host?

[UnLimiTeD]

Now, if the hardware you connect through is bad (few scattered nodes passing on data), that's a horrible noise penalty.
It might also mean a specific grid is not available.
You do not get a penalty to the grid your on just because of a temporarily bad connection, though, that'd be just noise.
As for wired nodes:
Bar turning off the wireless, you can put your server into a signal blocking box and have it only accessible by a throwback terminal that may(or not) be plugged into with other devices.
Means when your Rigger goes to sleep, the data is save unless someone crashes into the building and carves up the server housing to grab the harddrives (use autodestruct-triggers).

[Namikaze]

So to be clear, if the grid you are currently on goes down, you would also lose connection, yes? As would any intermediary grid between you and the destination grid?

This doesn't happen with the Matrix.  There is enough redundancy that the whole thing can stay afloat indefinitely.  At least that's the theory.  After the Jormungandr attack, the corps were looking for a perfect solution.  They developed the mesh network design from 4th edition.  Then they realized they made the whole thing too hackable, so they went with a median solution that is found in 5th edition.

Also, if I connected to a fantastic grid from a lousy one, shouldn't my traffic still be capped to the lowest common denominator (i.e. the lousy grid)?

You are talking about working across grids.  Let's take that example from before.  You're on the Seattle local grid and want to access the Ares host on the Ares grid.  If you don't move to the Ares grid, you get a -2 penalty to your actions while on the Ares grid.  This would mostly apply to hacking attempts, as legal actions don't require tests.  It's an abstract way to represent the difficulty and latency that comes with moving data across two sources that aren't normally linked.

It's hard for me to design the matrix component of runs as a GM. Even something as simple as the runners breaking into a small office and collecting data from a computer leads to all sorts of odd questions. Is the computer also a host? Is it accessible from the grid? Is it only accessible when you're close to it? Wouldn't that require it's own transmitter/receiver (basically, wi-fi) and it would kind of be its own grid? I'm not sure if that makes any sense, but it's actually the really mundane nuts and bolts I'm having trouble with.

I'm actually a web/software developer for a living so I tend to see all things from that perspective.

Forget everything you know and focus on the little bits of the Matrix that we do know about.  As a system administrator, I know the feeling of looking at the Matrix rules and wanting to pull my eyelids off in frustration.  The solution was to simply acknowledge that the Matrix is basically a totally fantastic land of make believe.  It does not make sense at all, and it might not ever.  When designing Matrix assets for my opposition forces, here's what I do.

Design the host
Design the spider
Determine if the spider is constantly attending the host

Host notes are written like this:

Host Name, device rating
Attack X, Sleaze X, Data Processing X, Firewall X
IC: <list of IC>

If the spider is attending the host, I run it like an NPC would do.  If the spider is not attending, or working from off-site, I usually give the IC a chance before bringing in the spider to mop things up.  I always run Patrol IC at all times.  If the Patrol IC suspects something is wrong, it usually triggers a low-level alert.  This may or may not alert the spider, depending on the owner of the host in question.  If the Patrol IC knows that something has happened, such as from being attacked by a hacker, it will bring in the big guns.  It triggers a system-wide alert, which will alert the spider and begin launching IC.  I usually run an instance of Binder and Black IC in that scenario.

Let's imagine another example, like a bullet train traveling through a remote area. The train can be controlled from the Matrix, and passengers can also use the train's "wi-fi" to connect to more distant grids. How would this be implemented?

Is the train a host? Is the train a grid? Is there a computer on this train that I could destroy to "destroy" the host?

The train most likely is neither a host or a grid.  The grids are everywhere at all times.  If nothing else, there are satellite uplinks that allow connectivity to the Matrix through magical means.    So the people on the bullet train can access the Matrix simply by connecting their device to the grids.  Once you stop looking at all the technical holes in the Matrix rules (you could drive a Roadmaster through most of them) the whole thing is really rather simple and almost elegant.

It's a shame that it's not more technically feasible though - but then, back in the early days of Star Trek communicators were practically magic too.

[Darzil]

Am I right in considering that essentially a spam zone reduces your matrix connection (causes noise) by reducing the 'bandwidth' available to you, whereas a static zone reduces the 'bandwidth' available to everyone?

I would presume it'd be 'realistic' to consider the noise level for a matrix action as being the maximum of the noise at the source and the noise at the target. Due to the mesh nature of the matrix we can (usually) ignore the route between the two. Considering just the source area would be a little unbalanced, making it almost always best to be sat in an ivory tower with perfect comms, considering just target area would make it almost always best to be sat in the middle of nowhere. Adding the two together would seem to be reducing bandwidth much further than that available.

[firebug]

In Splintered State, Brackhaven Investments HQ have 2 hosts.One is accessible wirelessly (security host) and one is accessible only with a direct connection (data host).So I Imagine a small company would do the same thing.A small server room with a private host.

Not an expert on the Matrix or  rl computers though :/

Sadly that doesn't actually work once logic is applied to it.  The pre-written stuff in Shadowrun is often plagued with stuff that doesn't actually work, in my experience.  Nobody's perfect I suppose.  The issue is, even if they do have a physical server that can only be accessed by standing next to it and plugging in...  That's worthless.

It can't do anything and it's about as useful USB drive chained to a desk.  If it was connected to anything else, you could use that to enter it (that's how accessing a wired network in Shadowrun works) and since it's explicitly not, it has no way to contact anyone or anything.  Again, if it was connected to anything else (which would be required for it to alert anyone about things) then you could use that connection to breech it.  If it's just a digital lockbox in the basement, once a hacker is plugged in, they can go to town on it.  It can launch all the IC it wants--  A decent decker with Fork and Hammer will probably be able to crush the IC faster than it will be launched, and doesn't have to worry about being discovered.  Once he has whatever info he wants, too, he just resets his deck to wipe his overwatch score and leaves.

TL;DR: Cutting a device off from all communication can actually make it much more vulnerable.

A paper book written in a cypher and kept in a nightstand would probably be a better way to keep information hidden.

[Sabato Kuroi]

In Splintered State, Brackhaven Investments HQ have 2 hosts.One is accessible wirelessly (security host) and one is accessible only with a direct connection (data host).So I Imagine a small company would do the same thing.A small server room with a private host.

Not an expert on the Matrix or  rl computers though :/

Sadly that doesn't actually work once logic is applied to it.  The pre-written stuff in Shadowrun is often plagued with stuff that doesn't actually work, in my experience.  Nobody's perfect I suppose.  The issue is, even if they do have a physical server that can only be accessed by standing next to it and plugging in...  That's worthless.

It can't do anything and it's about as useful USB drive chained to a desk.  If it was connected to anything else, you could use that to enter it (that's how accessing a wired network in Shadowrun works) and since it's explicitly not, it has no way to contact anyone or anything.  Again, if it was connected to anything else (which would be required for it to alert anyone about things) then you could use that connection to breech it.  If it's just a digital lockbox in the basement, once a hacker is plugged in, they can go to town on it.  It can launch all the IC it wants--  A decent decker with Fork and Hammer will probably be able to crush the IC faster than it will be launched, and doesn't have to worry about being discovered.  Once he has whatever info he wants, too, he just resets his deck to wipe his overwatch score and leaves.

TL;DR: Cutting a device off from all communication can actually make it much more vulnerable.

A paper book written in a cypher and kept in a nightstand would probably be a better way to keep information hidden.

I thought the data server in Splintered State had the purpose you describe : It's only there to store data,and workds as a digital lockbox that has more than enough physical protection ( guards,sensors etc).And IMHO it worked perfectly.At least it did in my table.

So cutting a device off from all communication makes it  more vulnerable once the decker has access to its physical location.Which is not an easy task for the decker.That's why the greatest banks on the planet are  in orbit , right?(At least I think they are, I may be wrong )

[UnLimiTeD]

Well, you can have it connected by wire to a high rating host with a spider, so the device is accessible through the matrix, but dare trying to get at it.
Then, a properly walled structure probably has a noise rating of +1, which means to have decent chances of hacking, the attacker needs to be reasonably close, which is what the corps want, so they can shoot them better once they find them.
You can also completely seal off information for the night so that a remote rigger (some third party security service) to be called by the night guards when needed is all that's necessary, saving on personnel costs for sentitive but not absolutely crucial data; If you don't need it for those 10 hours, no need to pay for active security.

[Namikaze]

Am I right in considering that essentially a spam zone reduces your matrix connection (causes noise) by reducing the 'bandwidth' available to you, whereas a static zone reduces the 'bandwidth' available to everyone?

I would presume it'd be 'realistic' to consider the noise level for a matrix action as being the maximum of the noise at the source and the noise at the target. Due to the mesh nature of the matrix we can (usually) ignore the route between the two. Considering just the source area would be a little unbalanced, making it almost always best to be sat in an ivory tower with perfect comms, considering just target area would make it almost always best to be sat in the middle of nowhere. Adding the two together would seem to be reducing bandwidth much further than that available.

Basically yes - it's all about how you want to interpret the Noise rating.  Personally, I see it as too much traffic clogging the pipes so to speak.  The problem is when you start to consider the jammers also create Noise rating, but don't actually bock communications.  According to Aaron, jammers block wireless bonuses.  *eyeroll*

I'm of the opinion that Noise is Noise.  If you have more Noise than you have Device Rating then you lose your connection to the Matrix, which kills your wireless bonuses too.

[Sengir]

Also, if I connected to a fantastic grid from a lousy one, shouldn't my traffic still be capped to the lowest common denominator (i.e. the lousy grid)?

No, because grids have nothing to do with the things of the same name from older editions which people keep bringing up: A grid is an set of devices with the only function that two devices on different grids get a -2 when interacting with each other. Also note that all grids are global, there is nothing stopping you from jumping straight into the Emerald City grid (Seattle) from the South Pole. Location, routing, and anything else you might associate with old telephone grids (or pre-SR4 matrix) do not play any role, it's a purely artificial subdivision of devices.