In every edition, Matrix rules tend to get a lot of flak; moreso perhaps than many other rulesets, and not without reason. There is a very good reason for this, however, and it has to do with the history and the development of the Matrix rules.

If you take a look at the rules for magic through all four editions, you can see that while a lot has changed, the basic principles and concepts behind the rules are consistent throughout the editions. First edition has spells, spirits, adepts and foci, and so has every subsequent edition. What that boils down to is Shadowrun has had four editions to expand, clarify, and revise the magic rules. Even with the significant system changes in SR4, a clear line of descent for something as simple as astral signatures can clearly be traced back to its origins.

The Matrix rules, unfortunately, are another kettle of fish all together. It isn't uncommon for Matrix rules to undergo significant revision, with much less "rules spread" (new rules produced in different books, rather than "core supplements" like The Grimoire) than magic - it was far more common to see a new spell printed in a supplement than a new program. That means that for every edition, the Matrix rules saw significantly less development when compared to other rulesets; by the time third edition rolled around the magic rules had two solid editions of development to work out the kinks, and the majority of the problems dealt with the newer material being introduced.

In addition, whereas magic is based mostly on the whims and logical structure of the game designers, most players and designers prefer Matrix rules to at least nod toward modern computing - which has been and continues to be in a state of constant flux and advancement. Hence the occasional appearance of apparent anachronisms in the Matrix: it is very hard for anyone, even professionals at the bleeding edge of their industry, to make any sort of prediction that holds true for five to ten years in the future, much less sixty-and game designers are rarely professionals at the bleeding edge.

Uh, yeah, whatever.

The problem is that the SR3 Matrix rules worked for WANs as well as LANs, whereas the SR4 Matrix rules are designed for PANs and already start to fall short on LANs.

The interesting part about that is the first reply, while critical of AH's statement, goes to his final point. The problem with th ematrix is someone comes along throwing around TLA's and their computing experience and tries to make it fit.

That's what I dislike aboput the Matrix rules, no matter how much you try and model real computers to appease the people who will try and impose their real world computer knowledge on the game, you will never truely satisfy them.

The only prediction I'm willing to make about 60 years in the future computing is that anyone here's predictions are probably wrong.

The problem is that computers exist in RL, whereas Magic doesn't. People are willing to accept a level of abstraction and a little "hand waving" in the Magic system because they don't have much in RL to compare it against.

The Matrix, on the other hand, is fairly analogous to a RL system and players are constantly trying to "imprint" RL technology and concepts onto the game system. I think to keep from running round needlessly trying to turn Shadowrun into some sort of computer dice simulator you need to detach SR computing and the Matrix from any relation to RL computing. Only then can you look at the Matrix rules from a rational "game balance" perspective.

Example: I haven't seen any question the arbitrary limit on bound spirits, yet there are constantly raging debates about "Agent armies" and such. The real solution is for people to just stop worrying about it and focus on the fun.

EDIT: Demerzel posted before I finished, but he is basically saying what I am saying. Computing is advancing so fast, I think everyone would be better off assuming that technology 60 years from now bears no resemblance to current technology, and move on.

as a strong supporter and critic of both the SR3 and the SR4 Matrix systems, i have to disagree, at least for myself. i'm an IT professional, and while i'm certainly not unaware of the glaring inconsistencies between any version of the Matrix and real-world computing, both versions are fun enough that i don't mind handwaving. both, however, have serious balance and internal consistency issues.

That's why all I ask of Matrix rules is three things: that they be logical, interesting, and not make my head hurt.

Realism isn't on that list.

I think that the current matrix rules, with some tweaks here and there, mostly satisfy my criteria. They ain't perfect, but I'm not bothered. I might be if I ran a hacker-heavy game (or at least if more players focused on the matrix). But then, I might also care about a lot of the arguments about how the astral plane works if we had more than one mage. It all comes down to priorities. A game like shadowrun is difficulty from a design standpoint, because you have all these very different concepts that need fleshed out, and some of them might get ignored or at least given little focus in many games. What one guy says is sufficient rules, another might find to be woefully lacking.

Honestly I'm confused why people try and compare RL computing and Shadowrun Matrix. Yes it's our world and it's 70 years in future but it's not "our" future the events in Shadowrun universe went along an entirely other line of cultural and technological advancement not to mention the world shattering events that have taken place in last 60 years vitas,goblinization,world wide riots,two major worldwide computer crashes etc...

So the 1st question is do I like the fictional conclusion of whats happened in this "other" universe or not.

Second question can this version of the game make it any easier or just add another version of confusing and obviously fought over rules.

As for first I'm good with what SR world has given me an continue to have a hell of good time playing in and living in this "other" world

As for second question, so far I've been pleased with 4th edition more so then I was with 3rd and looking forward with not a just little bit of optimism to seeing unwired and hoping it will make sense of if not clear up debates raging on the boards. Maybe even give us a version not unlike the magic rules of past editions as Ancient mentioned in his post.

I think a fundamental difference between 3rd and 4th is that so much of the Matrix rules in 4th are left up to each individual group and as a result the rules are kind of "fuzzy." In earlier editions almost every single action that could be taken on the matrix was fleshed out and given concrete rules. Now a lot of the stuff requires some thought and an understanding of how the basic rules work.

There's almost been a complete reversal in how to access the rules. When I started no one played the matrix because there were SO many rules. It wasn't easily accessible as a result. Once you figured out how the basic rules worked it was easy to play out the run quickly but it took time to get to that comfort level. Now the basic rules are spelled out quite clearly but not in great detail and the "learning curve" consists of figuring out how your group wants to play it and fleshing out the rules.

Personally I find the new version easier and more fun to play (although I did enjoy 3rd). There's a lot more that you can do now and the rules give the flexibility to do so.

Ignoring the whole "realism" and "suspension of disbelief" part of the matrix rules, the SR4 approach of turning hackers into script kiddies, where a 4 IP Sammie with skillwires running AR can be damn near as good as a dedicated logic 7 hacker running hot sim (and can kill him in cyber-combat while taking no damage) seems less than ideal.

What? Magic isn't real?!?!?

Oh, and combat also exists in RL, and nobody complains about those rules...

oh, wait... right...

Dread Polack

I'm going to disagree with most of what has been posted here so far. I think the fundamental problem with the Matrix rules is that this section of the book was cut too lean and it needs a few clarifying lines here and there and lost more examples.

I have some experience in the IT field and I live with a very experienced software engineer (C,C++ device driver type as he's always emphasising). I think I can justify a lot of the behaviour of the Matrix in 2070 in terms of actual real life extrapolation. It was written in one burst of creativity and could use a good editing and development, but I did a pure flavour piece on the Matrix here and I could expand on it. The one thing that seems hard to make realistic is the lack of lag in communication at great distance, I can't justify that. But everything else can be reasonably supported.

But in the rules themselves, it's best to keep a certain level of abstraction as has been done. That paradoxically protects our belief in the Matrix because it's harder to make the system completely internally consistent the more detail you get into.

I definitely agree that the transformation of hackers into script kiddies is a horrible mistake, though. That's why some people use tweaks like capping hits on Prog + Skill by Logic. It doesn't affect things much but it preserve flavour by preventing stupid super hackers.

If anyone thinks that the behaviour of the Matrix cannot be justified in plausible terms as I'm saying, then I invite you to respond not with simple disagreement, but with actual elements of behaviour from the rules and demand that I make them believable. (I ask only that I don't get me a dozen examples all at once.)

-K.

There's one thing I don't get. People complain that 4th edition turned hackers into script kiddies, but haven't they always been? Weren't the dice pools Skill + Program in previous editions, too?

(This is by no means saying that it's a good or bad thing. I'm just curious why this phenomenon is worse in 4th edition. Maybe I missed something, I'm a million times more familiar with 4th edition than the previous ones.)

well, you also had hacking pool in 3rd that was attribute based, or atleast attribute +skill, i forget exactly

the biggest complaint in earlier versions was that deckers had their wholly own world. sure the magicians had it to, but they could also do stuff in the mundane world. not so with deckers.

now they are released from this limitation, and turn into what can best be described as digital magicians.

as for turning them into script kiddies. even the big guys use programs. the big diff is, did you write it yourself, or did you grab it of some site and learn how to use it.

in a way whats needed is SOTA rules. as in, how to stay on top of the FTL rollercoaster thats high technology. what skills are needed to maintain a top of the line set of programs on your own.

then its like how rotbart von dainig put it, we have info about how to deal with the "home computer" that people own, and drones, but the rules right now cant handle large corp nodes with big number of agents and users running around. that is outside of splitting it up like how it was of old, with multiple "rooms" with special functions and their own ratings. if so, the rules have gone full circle...

or maybe one can do so, but just setting a overall device rating for the whole "house". and only specifying those parts that need exta ice. then define a kind of standard ice for the rest.

hmm, i need to think about that...

The only thing I don't like about Logic + Skill (limit Program Rating) is that it devalues higher-rating programs. If I choose to toss a rating 7 program into the mix, to represent the bleeding-edge of technology, it doesn't mean anything. People aren't likely to get 7 hits (in order to average 7 hits, you'd have to be rolling 21 dice).

Skill + Program (limit Logic) is slightly better, as it represents the hacker not being smart enough to use a program, but the same problem occurs: Nobody is going to be motivated to get another point of Logic.

There's only one permutation of this type left: Program + Logic (limit Skill). This is interesting, but has the same problem as the previous two. I really don't know what the implications of this would be, but I have to say it doesn't make logical sense to me.

All in all, I'd have to say I'm in the small contingent that likes Logic + Skill + Program Rating. This makes it so that both matter. Sure, you have to change the thresholds a bit, but there's a thread around here somewhere that has the conversions (I think it was either FrankTrollMan or DocFunk). I think this makes it the most consistent with the rest of the system, as it consists of Attribute + Skill + Equipment.

I would suggest a different fix, though it requires a bit more work: roll Attribute + Skill (I think Intuition should apply to some things, rather than always Logic), and each program provides a different benefit. Attack programs determine the DV (this weakens the program some, but it's not going to make people want it less). Edit programs add to the dice pool (requiring thresholds to be slightly higher). Medic programs add to the dice pool (but the roll has threshold 2, like first aid). And so on. Many programs will still add dice. Some programs will be thresholds for some things (Stealth would be the threshold to be detected by any matrix perception test). It'd require a bit of work (granted, more than most would want to put into it), but I think this'd be the best solution.

I like the 4th edition rules, as it makes the hacker able to be there on site as the mission goes on. It's not one player getting sole attention for long periods of time nor the hacker finding that their skill points are never used.

As for the fit between the real world and the game world, it works well for me. The game takes the basic notions of computing and fits it to an ordered and action-based combat system pretty effectively. It meets the basic-fit test, at least. Having it exactly mirror reality isn't something I'm looking for in a game (if I need that, I'll play GURPS old west where getting shot with an arrow means a surgury performed by the town vet and a week of roleplaying bed rest).

Cheers,

Andrew

I think they just need some examples of what systems should look like. Defences can range from wimpy to crazy with the same toolset. If they just did up some corp offices and said "typical defense looked like this' and released it as a web supplement or whatever, that would be grate.

From what I can tell is that they simplified everything and made everything have roughly the same rules feel to make facilitation easier from a player and GM standpoint for learning. Some people like other belly ache. The great thing is that rules are a guideline, something to go by. You should feel free to change them as necessary to fit your or your groups playing style. If you like 3rd, then integrate some of those rules in or substitute them all together. If you are going to cry about how hard it would be to do that then stop crying because no one wants to hear you gripe just to gripe.

The matrix rules have always been "special", since I started to play in SR2 times.

SR2: Much to complex for the GM, much to expensive Nuyen-wise for the players
SR3: A demonstration of the problems of variable target numbers, still to expensive
SR3: So far not a complete set of rules, and already a few hardcoded problems.

It is quite likely that the lack of use in previous editions resulted in a much lower speed of innovation. The magic book is the first rulebook published for a reason.

The rules got better (with every edition IMO), but now everyone has to use wireless tech, starting with breakfeast. Everyone who tries to do his matrix security finds that his comlink is script-kiddie territory and any communication must be considered unsafe. Several parts of the matrix rules did not survive the absense of practically unreachable target numbers.

A fundamental change in a rules concept invalidates most of the experience gained. Every change so far had merit, and the SR4 approach finally allows the decker to act together with the rest of the team. It´s just that nothing short of the feared IC-storm will RELIABLY stopp a hacker. And that is nearly as bad as a hacker that can do nothing.

There should be
- toasters that only process bread, not relay matrix data. The response rating of a toaster should be 0.1, not 3. It does neither need to run Doom XX on its spare cycles, nor a firewall.
- servers need to have firewall ratings in the lower twenties. Suddenly tactics like aquiring an existing password and hacking from a user account will be useful again rather than a waste of time.
- encryption strong enough to require storing of encrypted data for later decryption, not real-time-decoding like it is now.
- a limit to the ability of agents below that for hackers. the hacker should never prefer agents for convenience rather than skill
- all extented tests must be expected to be beaten every time. It might be possible to improve the extended test dynamic by introducing a threshold on each roll. Right now extended tests break both encryption and firewall tech.

Has it occurred to you that the purpose of Encrypt programs is meant primarily to be file encryption (and transmission encryption, if I'm not mistaken)? So no, nothing they do in the future MUST rely on those encryption rules, because those rules are primarily meant for file encryption.

PAN, WAN , LAN, Commlink, Device , it's all the same, it all boils down to a GM rolling some dice, and a player rolling some dice, so I want ti to resolve down to

GM Dice Pool versus Player Dice Pool
or
GM Network/Node versus Player Network/Node

I think it's a lot easier to visualize the abstraction level if you think of Nodes as a Network, not a single device.

Adding another machine to a network often doesn't make the network that much different.

p. 204
"Th e Matrix a complex organism, a vast collection of billions
of nodes all linked together in various networks that are
themselves linked together."

p. 208
"In order to enter some nodes (devices or networks), however—
especially private ones—you must actually log in to an
account."

p. 211
"In active
mode, you can both access and be accessed by other nodes
(PANs, devices, and networks)."

P. 216
"Node—Any device or network that can be accessed."

Substituting the word Network for Node in almost all text of the Wireless chapter makes a difference.

All this ticky tacky detail of N! Agents on N! Commlinks versus (N+1)! Agents on (N+1)! Commlinks duking it out on the matrix becomes Node X of Rating X versus Node Y of Rating Y. Much simpler math.

You can imagine adding one machine and one more copy of software to a Network Backbone doesn't really add much to that network in terms of it's ability to attack or defend against another Network.

In using the matrix rules I've found them to be pretty consistant and workable.

The big problem is that hacking gear is too accessible. Hackers end up with systems at character creation that are the 3rd edition equivilent to decks and programs that would were priced in the millions of nuyen.

A hacker with these systems can stomp all over anything that isn't rating 6. Worse there is no way to effectively keep a hacker out if you wanted to, short of turning off the node. Changing the mode of the node only delays the hacker by a few seconds. Encryption might as well not exist and once inside a hacker has access to everything.

I'll buy that. So decryption and spoof.

Matrix dungeon crawls are actually a good idea, since they provide an excuse for the decker to actually go with the party. Connecting directly to a target node would be easier than searching it through the dynamic wireless web.

Right now breaking a passkey is never required. I did not miss the few lines on those - a hacker can easily establish his own admin account on any system he chooses to. Thats why passkeys are worthless. Any key that does not depend on hardware can be set by the hacker (I would say that writing emulation software for hardware keys requires access to the key, so those do something).

Right now any file can be decrypted (hardware keys excepted) within seconds. The only usefull (and near failsafe) use of encryption is a high-level agent with high-level encryption software constantly reencrypting a PAN. That does give meaning to the PAN concept, IF all systems on the PAN can run decent encryption ratings (4+ is rather safe, even 3 will usually work).
Want secure communication? Set up a secondary comlink network with only the comlinks on the subscriber lists. One comlink runs an encryption agent, all others only encryption software. Devices that need connection to the outside world are still vulnerable, but neither combat drones nor cyberware need that.

It would be easier and better to make real-time decryption not an extended test, but one can make RAW work.

(The former ideas are for real. But has anyone considered running pilot instead of system on a comlink? Instant self-defense system with YOUR full load of programms at service and matrix combat-ability to boot? A friend of mine "enhanced" the idea by suggesting eyeball-drone-mods that are on your PAN anyway and already have pilot ratings...)

The way to set the interval for decryption would probably be best if it was based on the relative size/complexity of the file being decrypted. It would still be a matter of GM fiat to determine what interval to use, since there is no canon measurements of file size (rules-wise, that is), but a set of guidelines based on this would be good. Giant file containing complete scientific research data and designs for prototype device: interval 1 week. Text file transciption of phone call: interval 1 IP.

I wonder if encryption should come in grades, like it does in real life. Just change the price structure to an exponential curve and set the interval based on the rating.

Rating 1: cost=free with device, interval = 1IP
Rating 2: cost=cheap=100nuyen, interval = 1 turn
Rating 3: cost=average=1000nuyen, interval = 1 minute
Rating 4: cost=pricey=10000nuyen, interval = 1 hour
Rating 5: cost=expensive=50000nuyen, interval = 1 day
Rating 6: cost=atmospheric=100000nuyen, interval = 1 week

This way your runners and high-level fixers or Johnsons' communications and files are going to have an interval of an hour, maybe a day. A research facility or bank will have an interval of a week. Swat teams will have an interval of a minute. Security guards and other chumps will have an interval of 1 turn. Average wageslaves will have an interval of 1 IP.

You are talking about software - that will be pirated.

There is no easy fix by 'rating'. Personally, I would stay with 1 turn as basic Intervall... but double Intervall for every successive roll and use Rating² as Threshold.


Response 6 + Decryption 6 on average produces 4 Hits.

Encyryption 1 needs 1 Hit to be broken. This means 1 intervall on average: about 3 seconds, or 1 turn.
Encyryption 2 needs 4 Hit to be broken. This means 1 intervall on average: about 3 seconds, or 1 turn.
Encyryption 3 needs 9 Hits to be broken. This means 3 intervalls on average: about 21 seconds, or 7 turns.
Encyryption 4 needs 16 Hits to be broken. This means 4 intervalls on average: about 45 seconds, or 15 turns.
Encyryption 5 needs 25 Hits to be broken. This means 7 intervalls on average: about 6 minutes and 21 seconds, or 127 turns.
Encyryption 6 needs 36 Hits to be broken. This means 9 intervalls on average: about 25 minutes and 33 seconds, or 511 turns.