RunnerSmurf, what you have done is WAY more helpful!

My comments:
1. I like it. Skills are too "chunky" in SR4.
2. Good ideas, but simple is better.
3. I have had similar thoughts. More statistical research is needed.
4. Meh.
5. I agree, I have actually done MASSIVE research to add seating and entryway info to a spreadsheet I use for my campaign. Likewise concealability for weapons.
6. Right now, there is a mess between stat/skill bonuses and dicepool adders. ALL adders need to be mods, capped by BASE skill maybe as you suggest. This includes Specializations!
7. Or VR downgrades action length by 1 step, Complex becomes Simple, Simple becomes Free, Free becomes "no time at all"? This makes VR "as quick as thought"!
8. Concealability at least needs to come back!
9. I think, if nothing else, Drain REALLY needs to be revisited!
10. Or the damage penalty steps are in increments of 2: -2, -4, -6, etc.

I like this list. Here are some basic ideas as to how I'd like hacking to look:

Wireless Doesn't make a device insecure: Wireless encryption cannot be broken. The only way to hack a networked device is to either gain physical access to it, and thus the hacker must go with the team, or to go through the network, where the hacker can take the team with him (more on that later).

Spoofing an object is possible, but only if you've alreay aquired the encryption key. This can be done by:

a. Hacking the device through a physical port and finding the encryption key.

b. Opening the device and using special tools to read the encryption key directly from the device's memory.

c. Hacking the network itself to find the encryption key.

Once the key is acquired, it only requires an action to send a command to a compromised device.

Encryption on files can be broken, but it takes a long time and powerful hardware, so isn't done during the run.

Barrier to Entry: Being a hacker requires a cybernetic implant called a Cyberdeck, as well as other specialized computer equipment. It also requires skill. I'm not entirely convinced that programs should even enter into it, what with software piracy.

Hacking is a Team Effort: Cybercombat, unlike hacking, only requires a little skill and the right gear. If all the members of a team are networked through the Hacker, they can join him in cyberspace. When the Hacker gains access to a node, so to do his allies.

Essentially, the hacker does all the intrusion work, but the team can help in cybercombat so long as they have a few points in Cybercombat and a weapon.

Cybercombat weapons should be part hardware. Cybercombat is often handled by actual people helping a network's spider. This way, cybercombat really is combat. IC are similar to drones in that they require special hardware to partake in cybercomat and have to be plugged into the network.

Even Joe Public is Protected: Hacking used to be a problem for individuals and their commlinks. These days, Matrix Service Providers and independant companies offer security networks. For a monthly fee, you can have all connections routed through this network, which provides you high levels of protection.

Thoughts?

Skills

For Shadowrun 5th Edition, which will exist eventually one way or another I think they need to continue the trend they started of streamlining things. That's making things easily for both players and GMs alike. I liked and still like target numbers, don't get me wrong, but it's much simpler with static numbers because it allows the GM to just call a roll and the player(s) get to it and report results. This takes work away from a GM and makes their life a bit easier.

Skills being capped to six wasn't a good idea in my opinion. There needs to be a cap but it needs to be raised, but at the same time offer play styles, so someone can pick up a character sheet and see what kind of PC they have before them. Last year I worked on a system that ranged from 0-12 with an increased cap of 15 for folks with aptitude (+1 for aptitude was pathetic), that doesn't work so well however.

The SR4 skill system is nice that you can create an character from scratch that is along the elite at a chosen area, but as a player you're then stuck and cannot get any better, that's not ideal for anyone, GMs or players. The solution I think is to break down the specialisations and expand upon them and introduce new levels of specialisation (dedication).

Skill Redux - Shadowrun 4th Edition

The above is still WIP, but usable. I only wrote up the document this evening, but have been working on it and discussing it for a while now. It maybe full of typos too

My system raises the skill limits to 18 standard and 21 for aptitude, importantly though after the skill, you need to select a particular specialisation. This is nice as it allows players to really narrow down their characters angle. You see, in SR4 (and SR3 and SR2) two completely different PCs that require high stealth may actually have different skillsets. For example a black op specialist and a private investigator. The black op specialist may have stealth -> infiltration whilst the private investigator may have stealth -> stakeouts or perception. Depending on whether he prefers sitting in a car and looking at a joint or tossing one.

Virtually no complexity is added and it allows players to really get their characters where they want them without all having carbon copies.

For example. John the PI (and ex-lonestar officer) and Tomahawk the N.A.N. covert ops specialist would end up with different skills rather than the current situation where both of them could easily end up with stealth 4 or 5. John the PI could have Stealth 3, Stakeouts 4 and Tomahawk Stealth 2, Infiltration 6. That allows them to be different individuals. That would give John the PI 3 dice for all stealth checks but 7 dice for stakeouts (+attribute/mods) and Tomahawk 2 dice for perception/stakeouts but 8 dice for infiltration.

Skill groups were a nice idea, but I think they should basically contribute to the rest of the skill (as it were) and the "specialisation" should also range up.

That way people can make the "best of the best" at Chargen (and some expense), but don't actually need to and can grow more in game. This way you can keep the base SR4 descriptions of skill levels. When I changed skill caps to 12 (15 aptitude) I looked back on the ex-renraku black ops PC and realised that his 5 in automatics hardly made him near the best in the world, whilst he was supposed to be amongst them. This system sorts that out to a degree, because he's then been made as a generalist (and can then specialise in assault rifles).

I'll get back on combat, firearms, the matrix and general rules later.


TL;DR: Change the system to Skills -> Specialisations -> Dedications. Each ranges from 1-6 (or 1-7 with aptitude). All modifiers stack backwards. Pistols 3 -> Semi-automatics 4 -> Ares Predator IV 5 and using a Colt Manhunter? 7 dice. Ares Predator IV? 12 dice. Allows players to create unique PCs with the "same" skills and isn't much more complicated.



- Baat

How so? The GM still has to tell the players how many dice they roll. I don't see what makes "roll with -6 dice" more practical than "roll at TN 6".

Firearms and Combat

Combats not to bad as it is now. Although it doesn't feel as rewarding as SR3 was. You know, making that impossible TN? Instead now you potentially get your dice poll reduce to hell and, it just feels like a penalty more than anything else.

The first thing to do is introduce thresholds for certain things, much like Runner Smurf has posted about. I've briefly spoken to him and I basically did the same thing as he and it works nicely. Potentially it causes combat to become a longer affair and I don't particular think that it's a good thing, so weaponry should be tweaked as well.

Runner Smurf's Post and .PDF

By using thresholds you get back that SR3 feeling of, ".. oh my gawd, I need TN 14 or my buddy dies!?.... OMGWTFBBQ I DID IT!", which you don't really get with SR4. What Runner Smurf says in his post about average skilled riflemen hitting people a kilometer away is true and my beef as well. However with thresholds comes longer combat, to counter this you only have to up some numbers here and there (and then change a bunch of other stuff). I've been working on this also.

Combat Redux - Shadowrun 4th Edition

That document is completely unfinished and WIP, but I'm putting it forward now to give an example.

The nice thing about this is a damage chart and AP chart (with range etc) would only take up a page and allow people to recreate real life weaponry and make custom weaponry. Also it's more-so realistic (although I need to finish some sections about weaponry modifications for damage and such) than the book example. I think with Runner Smurfs efforts and the damages nothing ends up broken. More misses, more lethal when rounds enter. It would all depend on playstyle, although some of the typically "big gun"s aren't really that bad. Important is the Armour Penetration rule. This really creates the difference between high and low velocity weaponry.

Say an Ares Predator is 10mm (or .40-.44) and take an AK-47.

Ares Predator IV: 7P -1 AP.
AK-47: 5P -6 AP.

So the Ares Predator hurts. Yes. But against full body armour the Ares Predator requires 2 successes (remember the threshold deal) to deal full damage, otherwise you half it and it's stun. Where as the AK-47 reduces the armour down to 4 and deals full physical damage.

I think this makes the game more real, without too much complication and would also create a more cinematic experience of bullets flying everywhere and when one DOES connect, bits go flying all over the place.

The last point I'd make is from Runner Smurf's threshold system. With thresholds, you get to put excellent modifiers on certain devices and actions. For example Runner Smurf's system grants smartlinks a -1 threshold modifier which suddenly makes it beyond awesome again and smells totally of SR3 to me. Likewise my major beef, Melee, can get an easy boost by granting +1 thresholds for shooting in melee and +2 for shooting melee at someone attacking you. Melee combatants take a lot of risks closing grounds and that bravery (stupidity?) should be rewarded if they get there without being booked.

TL;DR: Use a threshold system and increase damage across the board with easy to see numbers. This should creating a more cinematic, but realistic experience.



- Baat

*awaiting flaming for unrealistic numbers*

Granted. Yet sometimes there are no modifiers and the GM can just wave his hand and the player gets on with it because they know what they're doing.

That aside. Without a big old dice, any TN system always creates imbalances, unless you using 3-6 (which I've mooted in the past but thought better of it). Either way, I feel that SR4 is simpler for new players (and definitely GMs) and that's very important to any RPG.

If we had a say in the creation of Shadowrun 5th Edition, the most important thing to remember is that it's not REALLY set for "our" needs, it's for new players, to allow them to ease into the game and enjoy it. That brings more players, more money and high quality productions, which in the end does effect us.



- Baat

I absolutely Hate variable TN's...

Sorry, had to do it...

I am amazed every time I hear someone reminiscing about needing to hit that ridiculous TN21 or whatever it was they needed. I am much happier with the current system.

As am I... Fixed TN's equal Happy Times.

I've heard a couple of complaints that Skill ratings capped at 6 is too low. Could someone explain why?

The Matrix

Okay.

I know this will be frowned upon to a degree. But I feel the Matrix needs a complete refresh. In all the games I've played Deckers (or hackers if you like) have been completely over looked and left to NPCs. Why? Because they require an almost solo run and leave players in limbo (whether the decker or "other guys") for time, which isn't great for a good gaming experience.

I think Shadowrun 5th needs to do what they really were with Shadowrun 1st and assume about the future and whilst doing so steamline everything. Decking is far too complicated and time consuming (and solo, did I mention that?). What needs to happen is that decking is transformed into a more cinematic experience that any GM will feel comfortable in running and players can relate to. Also, simplicity.

I've begun working on an unfinished set of rules to completely change the Matrix and how it works, jumping ahead of SR a generation or two, but I think ultimately it would lead to a better game where deckers can be used freely and more importantly, their meat head team mates can join them on matrix runs.

Matrix Redux - Shadowrun 4th Edition

The way I see it, the Matrix should be a obsidian black and white neon reconstruction of the real world. The Aztechnology pyramid is 375 levels? So is the matrix construct. I think the matrix should resemble the normal world, with exception of places without wifi or networks (like how many is that gonna be in 2070?). The way I see it is that people with commlinks on should create a ghost like shadow in the matrix, this way a decker can actually follow their team mates about in real time. They can stand in the same "room" as them, half being able to see them.

In a gameplay sense, this allows a GM to "include" a decker in the action, but adding in matrix elements, such as that guard by the door (a firewall?). I think combat should be resolved just like the real world and people should bleed the same (of course it's all matrix constructions). This would all be like The Matrix films (I swear that wasn't my basis). A decker should feel like their "there" and not a program.

I think with the A.N.I.M.U.S. idea (I swear, I just needed a good name, it's not based on that) being able to include team mates that the entire matrix scene would refresh. Imagine a decker getting into big trouble in a corporate compound because his team set off the alarms and is beset with dangerous IC and maybe an enemy decker, so the heavy weapon specialists jacks in to try to give him a hand. Sure he won't be able to do much other than shoot at stuff (and not as well as the decker), but I think it creates a good game atmosphere which is important and importantly keeps everyone included.

Primarily, the reason to make some changes is to simplify decking to existing standards, so that anyone can play a decker if they have a good idea for one. I really don't think decking should be exclusive to the masochistic GM that spent two hundred hours learning the rules.

Importantly and not mention in the .rtf is that the Matrix would have reality changes, mostly in gravity, but would run the same as normal. The weapons you arm yourself with work as the attack program, but you follow combat normally. So that a heavy machinegun does the same damage it would normally deal. This keeps everything on one level and importantly everyone rolls the same amount of dice they usually do (with a -2 of course).

TL;DR: The Matrix needs a full scale rework so that a GM is capable of running a game for the team and the decker at the same time and also including the rest of the team on matrix runs so that the decker actually gets to feel badass in combat once in a while.



- Baat

There's too little variability in the skill ratings. You go from 1 (barely trained) to 6 (the best in the world). The difference between those? 5 dice. In a game where dicepools ranging from 15-25 are not uncommon. Skills should be making up at least half of your dice pool, which means either increasing skill caps, or decreasing contribution from other sources, or preferably (to maintain roughly the same dicepool sizes cause we really don't need size 40 dicepools) both.

To give a example, right now we have a character in our group who is primarily a street sammy. He has 3 logic, and he picked up as a side skill Demolitions 5, so he could contribute outside of just shooting stuff. However despite being nearly one of the best skilled demolitions people in the world, he has 8 dice to roll. On the other hand, I am planning on introducing a logic based character into this campaign with 1-2 ranks in all of the logic based skills, and a really high logic. With a single point in demolitions, I have a dice pool of 13 or 14 to throw at demolitions tests. With a little more money I could crank that up to 15-18. Basically, as a pretty smart guy with almost no training, I tell the guy who has trained extensively in this field "Move over, I can figure out how to do this better than you", and that isn't right. Having high skills should mean more, and high attributes should mean less.

This is already happening at some tables. This is how our table actually runs, in fact. *shrug*

I thought I did, but I will again anyway.

You make a character.

He's an Ex-Tir Ghost Sniper.

You take aptitude (longarms), cap longarms at 7 and specialise in sniper rifles. You do a few sessions. You get a bucket of karma. You look over your sheet to make your sniper better at sniping, because that's what he does and that's what he's been doing. You realise you cannot get any better.

Maybe that's not a problem for you, but it is for me.



- Baat

He can still increase his Perception, Disguise (camo),... Only 1 skill can be maxed out at chargen. Your problem could also be solved by lowering the maximum skill allowed at chargen.

I messaged Runner Smurf a couple of days ago after seeing his Revised Combat Mechanics were almost identical to the ones I'd come up with (and was using).

I was suggesting some kind of joint effort to create a set of steamlined rules. If you have Matrix mechanics similar to what I'm suggesting, would you be interested?

I know mostly it would be a waste of time because mostly everyone would still play SR4 (a) core, but perhaps if they were good they would get integrated in to SR5 or in the least we'd all (that care) would have a refined set of rules to work by. I stand by the stance that rules cannot be made by one man alone and that multiple minds breeds greater creativity.



- Baat

He wants to get better at aiming with a sniper rifle. He cannot do that. He is limited by the rules.

As Seerow said. You get 1-9 from attributes and 1-7 from skill. Then mods like smartlinks and such. In the end a point here or there on skills barely makes a difference when you're playing with such high numbers.

The problem with skills being 1-6 and capped at 6, is that there is no advancement left for your primary pursuit. Yes, you can pick up secondary skills that enhance it indirectly, but that's not really the point is it. It's why adepts and magicians have a magic attribute. So they can get better at what they "do".



- Baat

Other Rules

Cyberware

Cyberware limits need to be removed I feel. Have racial maximums by all means, but I feel that the SR3 system was better that you could naturally pump your attributes by *1.5 and then cyber/bioware added to that. I see what the designers were trying to do, making cyber/bioware more important and that natural cannot be better than technological, but I see no need to cap the limits of cyberware. Secondly they wanted to keep everything within a certain dice pool level, which I understand, it's easier from a game balance perspective, but then if you open up the skill caps a bit...

To be honest, with the modifiers available, it's only possible to go so high anyway.

Magic

This won't be popular at all, but I think astral space needs to be removed or moved into metamagic territory. Too much paranoia and second guessing comes from the fact that you *could* always be followed and with the open/lethal/real world that Shadowrun is (unlike say D&D) it's easy to come to the assumption that the PCs are followed most of the time. I'd probably settle for slower astral movement speeds though or a way for mundanes to have a clue they're there. However spirits being only able to be "in their realm" or "in the real world" has a dramatic effect on the way the game works and I feel for the better.

As an aside, the astral space (and recon mostly) also creates that "solo run" kind of feel which really needs to get eliminated.

Hermetic and Shamanic traditions should be returned to what they did in SR3. I appreciate that the designers simplified the entire affair that way, but they both lost their feel. A quick fix is that only Hermetic's can bind and only Shamanic's can summon, perhaps with -1 to stats for "spirits" and +1 stats for "elementals". Yet it's far from perfect.




- Baat

How do you improve on Legendary? It was your choice to have a Skill at that level. What makes you think that it SHOULD be able to improve from there? Why not make him with a lower Skill level than 7, like maybe a 4, for a Veteran Level of ability? WHY does he have to have that 7?

Besides, as has been mentioned. Legendary in a single skill does not make a Legendary character in the requisite skills for the chosen filed/Specialty.

So What...

Maybe a character should be developed in a Way that makes a bit of sense. Hard to be "Legendary" in a series of 6 linked skills (you know, those that are required for the role you aqre playing) when you cannot buy the skills at that level. Why not have them all at 4 (which is still bad ass), and then increment into that legendary status organically. Doing so at Character Creation leaves a huge number of gaps that make no sense. This is where I htink the problem lies. The complaint is that there is no path to growth, and I call BS on that.

Why does 7 dice from skill equal legendary should be your question. Yes, the current fluff is that 7 in a skill is legendary. However, the crunch fails to support that. You can have a guy with 7 skill rolling fewer dice than a guy with 1, or even 0 skill, with better attributes and modifiers.

If you want to get stuck on "7 skill is legendary" as it is currently then yes, asking to go past 7 is bad. However if you consider an alternative where skill caps are significantly higher, then 7 is no longer legendary, but just pretty skilled. Something like 12-15 might be legendary instead. This both opens up further skill advancement post character creation, and makes someone who has his skill increased to legendary levels actually have a dice pool that outclasses others who are amateurs or barely skilled. Someone with a higher attribute or better modifiers might be able to match his dice pool at a lower skill level, but to pass him you have to have those things plus the same skill level.

It is not a question for me. That is what the fluff sets it as. All things being equal (and it can be) the Skill 7 character is far and away better at the skill than the Skill 1 Character. The Crunch DOES support it. The poroblem you fail to account for is that Shadowrun lets you bypass the mundane levels of accomplishment by allowing you to boost your attributes beyond what is normally capable. There SHOULD be hard caps for Skills. It makes the world more real. I am happy with that cap being a 9 point Skill rating system, which we already have in place. Unskilled, 0, 1-7. Why do you need more?

And, as normal, My question to you would then be: Why do you feel that you NEED skill caps higher than 7? All you are doing is creating additional Dice Bloat, which is a bad thing, in my opinion. As for bypassing the guy with Legendary abilities; it is already the way you describe. ALL THINGS BEING EQUAL, which is what you are saying here, you will need that Skill of 7 to be legendary, while the guy with everything and a skill of only 1 or 2, will still be worlds aways from Jow Blow, though he is still NOT legendary. You will always have characters that will roll less dice than a Skill 0-2 character with everything going his way, even some in the Best of the Best/Legendary range. That will not change, as long as there are mechanics to be exploited, and they will ALWAYS exist.

I don't need a skill of 15 to feel useful. Hell, most of my characters don't even have skills of 5 and 6 in their primary Fields. And once again, maybe that is where the problem lies. Not in the way the world currently works, but in the interpretation of how the world currently works, from your POV. When the Primary Elite Opponent (Tir Ghosts) is only slinging 17 Dice (up to 21 with a Tacnet), why are you trying to outdo them at character creation? You see, I am happy with that comparison. And I don't try.

There are a ton of ways to improve on a character that has a 7 in one skill (not sure why you would want to start someone there in the first place, but whatever). The initial methods involve improving those skills that are direct support skills, which will only be at a 4 (or even less) at that point (it is hard to be a legendary Sniper, after all, if your Stealth, or outdoor, skills are crap). I have yet to see any character I have created with zero potential for improvement, even the one or two who are beyond exceptional to start with. I am just curious as to why you do not see that yourself.

Ehhh... Rant Over... Apologies.

I agree, we don't need more dicepool bloat, and that 15-20 dice is more than enough. What I disagree with is that skills should only contribute 25-30% of that dicepool. I think skill should be the driving force in character actions, with attributes and other modifiers taking a back seat. Right from the beginning my suggestion was higher skill caps, but attributes giving only 50% of their value to dice pool, and having a hard cap on dicepool modifiers, dependent on either skill or attribute.

So instead of having your guy with 9 agility, 7 skill, 2 specialization, and 2 smartlink rolling 20 dice, you have agility 5, skill 12, smartlink 2, specialization 2, giving you 21 dice. ie nearly the same dice pool, but it requires a heavier investment in the skill to get that dicepool. You can't just get great attributes and be incredibly awesome at everything, the game becomes more about your skillset and less about how awesome your superhuman attributes are.

What we do for skills is they are now capped at 9, but it costs double to raise past 6.

Regardless of how you do it, if it's a flat cost per skill level and a skill cap, then it's a race to the skill cap and you're back to not improving anymore. It doesn't matter how you move the goalposts, it's still the same game.

Except it does change the mechanics to match the fluff (ie high skill really matters, and people with really high skills are legendary), and deflates the dice pools a bit longer, because even with a flat skill cost most people can't afford the skill that high. Right now anyone can start a character with a 8-10+ in their primary stat and 6 in their primary skill, because of the easy availability of stat augmentations. With the change, your base dicepools are going to be a good 3-5 lower in general, even if your first few missions worth of karma are going into racing for skill cap.

Another alternative is instead of a flat skill cost, but not a scaling skill cost as it is now either. Instead make it a step progression, like most nuyen costs run. ie 1-3 in a skill only costs 2, 4-6 costs 4, 7-9 costs 6, 10-12 costs 8.

That way it still costs an investment of 6+12+18+24 = 60 karma to max out a skill, which is pretty substantial. Currently that investment is 2+4+6+8+10+12 = 42 (or by BP it's 24 BP which is roughly 48 karma), plus another 34 karma for aptitude+rank 7 (though that combo is really not worth it. aptitude should either give a bigger bonus or cost less to pick up. Holy crap you pay as much for that 1 extra die as you do for almost all the rest put together!).

As long as the cost is scaling, you can just cut off the hard cap, leaving it to each person to determine where it's no longer worth it for them to continue investing. But personally I think 12 is high enough to give a good variability.


edit: So I was thinking and had a few more ideas to flesh this out a bit more:

-Base skill cap is 9. Costs increase on a step basis as indicated above with costs slightly increased. (1-3 is 3 karma per, 4-6 is 5 karma per, 7-9 is 7 karma per). Maxing this out costs 45 karma, pretty close in investment to what it costs now to get rating 6.
-Dicepool bonuses are limited to 1 per 3 skill ranks. So someone with 3 Firearms can get at most +1 extra dice pool mod, someone with 9 firearms can get up to +3. Specialization can raise this cap by an additional +1 within the appropriate specialization. (So the guy with Firearms 9 specialized Heavy Pistols can have +4 dicepool mods while using his Predator)
-Attribute applies at 1/2 value, as mentioned previously
-Aptitude increases max skill value by 3, and reduces karma cost of increasing the skill by 1 per rank. This helps offset the high cost of aptitude, making the total package of aptitude + rank 12 skill cost 80 karma, vs the 76 karma it costs now.


The result? A character with 12 skill really is amazing at using that skill compared to others. A guy with 9 attribute and 4 skill is getting at most 12 dice in his pool, the guy with 4 attribute but 12 skill has a maximum of 21 dice in his pool. If the guy instead had the max 9 agility, he could have as high as 24 dice in his pool, and that's as high as anyone can get in their pool pre racial mods/genetic manipulation. But even with fiddling with all of that you'd probably net at most an extra die or 2, making your hard cap of dice pool around 25-26. Most dice pools however will average closer to the 10-15 range as intended. Skills are generally cheaper to purchase per rank, but are more important so you want more of them, as you can't just rely on huge stats to drive your dicepools up to high levels.


Edit2: Realized that the above makes teamwork tests nearly useless. They would be another exception to the dice pool modifier limit.

I can see the case for skill ratings above 6. It gives sammies something to sink karma in while the mages initiate, which attacks one half of the "mages-need-karma, sammies-need-money" divergence.

But if reaching the maximum too easily is the problem, then make very sure you can't start out at too high a skill rating. From your maximum starting rating to the eventual maximum should be several, karma-expensive steps.

This is just a thought, and might not work everywhere, but what if dice from an attribute were limited to the rank of the skill?

Thus, skills allow you to better take advantage of your superior attributes. This makes the most sense for Agility, Reaction, Intuition, and Logic based skills. Less so for the other attributes.

Oh man.

Defaulting.

Actually, if you don't mind obscene numbers of dice (evil grin),

You get attribute TIMES skill dice, plus one if the skill can be defaulted, plus zero if it cannot.

(only mostly a joke. But I'm seeing so many fun variations I might as well stick that one in too.)

Yeah, I see the problem with defaulting.

Let's look at the three issues at hand:

1. Skill representing a small percentage of the dice pool.

2. Skill caps leaving no room for character improvement.

3. Character with high attributes but low skill being equal to someone with high skill but low attributes.


The first problem is really a side effect of extra large dice pools. Augmentations, gear bonuses, and conditional modifiers can great absurd dice pools. Locking it at twenty feels kind of like a hack, and Attribute+Skill x2 means that a full half of the dice pool has nothing to do with your character.

The solution is to let skill also represent how well a character can negate penalties and take advantage of bonuses. Thus, total modifiers cannot exceed the character's skill.

Question: should this limitation include extra dice from augmented attributes and skills?


The second problem shouldn't be solved. There should be a maximum skill level. People simply can't get infinitely better. Perhaps, though, caps should be put on character generation.


The third problem is best prevented by factoring attributes into the cost of improving a skill. Having a high logic or agility makes it easier to learn related skills. If improving an active skill costs 3 × New Rank − Attribute. It solves the problem by eliminating characters with low attributes buy high skills.

In this, I disagree. I do not believe there should be a maximum skill level. The weakness is that sooner or later everyone reaches it, and once everyone is special no one is special. Caps are a mencken solution.

I would propose instead some means of diminishing returns. It is either more difficult to get the 20th die compared to the 19th, or the 20th has less impact. My experience leads me to believe the latter method is better though much harder to implement. The reason is simple: the person with 19 dice is more likely to be earning more points (be they karma or experience points or whatever) than the person with only five. As a result the actual difficulty for both five to six and 19 to 20 is subjectively identical.

Again, however, I roundly object to arbitrary caps. They are outstanding examples of solutions that are obvious, simple, and wrong.

First, I'm going to say I haven't seen any reason why my suggestion (first post on this page) wouldn't work. It results in roughly the same maximum dice pool, an average dice pool closer to the average the developers expected, and makes skill a significant driving portion of the dice pool rather than an afterthought.

But here's my comment on Trillinon's suggestion:



Augmented Attribute is a part of the attribute. But augmented skills (such as reflex recorders or improved ability) should just be dice pool modifiers, not actual increased skill.



This actually makes the problem worse. It makes having a high attribute not just beneficial to the skill, but absolutely required before you start buying the skill. It also means you get the first 3 ranks of any skill you have a high attribute in for free, which is pretty absurd. Remember, the problem is that the fluff says having high skill ratings means something, and in actuality it doesn't. The problem isn't high skill low attribute characters with decent dice pools, but with high attribute low skill characters being better than highly skilled characters. This 'solution' makes it a little better only by virtue that someone with a high attribute won't have low skills, since the first several ranks are free. It doesn't change that attribute is the most important part of a character's success rather than skill.

You are not a beautiful and unique snowflake.

Let's take a look how others RPG's deal with character evolution (emphasis on skills):

D20: each level you gain a flat amount of skill points according to class and the maximum amount of points you may have in each skill is limited by your level. If you take into account that campaigns will stop at level 20 and not enter into epic territory, then everyone eventually reaches the maximum possible skill if they want so.

GURPS: per se, there is no maximum skill level, although, at higher levels (18+) it is only so good to counter-effect any possible penalties, also, it gets more and more difficult to increase each new skill level when you are already in the higher levels of it.

L5R: it never emphasizes any maximum level, and you can increase as high as your GM allows it. Also, it sets very clearly that in no way a begining character may start with a skill level higher than 4.

New World of Darkness: the maximum normal level is 5 possibly getting higher according to your supernatural attribute (Primal Rage, Blood Potency, etc.), although it doesn't prohibit it, NWoD makes it harder to begin with the maximum skill level possible.

So, it's only natural that there should be a maximum skill level, it should just be so that begining characters should not already be at the maximum possible level and should have room to grow.

Are you aware three of the four examples you chose support my position?

I'd argue 2 of 4. GURPS and L5R support (neither has a cap), D20 and NWoD both have a cap, though NWoDs cap can be raised, there still is one.



Going capless is possible, as shown by those examples, the problem is figuring out the sweet spot where you want most players, figuring out how to make it cost effective to get to or close to that point, while making it undesirably expensive to go much further beyond it (but still possible if you really want to make yourself completely one note).

Diminishing returns could work, but if we're going for something at all simulationist, then those returns diminish very fast at the higher levels. It 6 or 7 is supposed to represent humanities finest, then even getting there should take a lifetime of training. Getting better than a 7 should require decades to achieve, all the while your body degenerates.

In truth, I like the idea of building a system like this, but it ultimately has the same effect as a cap. People can only get so good at something because time and mortality are against them.

The numbers would have to be adjusted. We don't want free skills, but the concept is still sound. It solves the problem by removing one of the cases, just not the case you wanted removed. You see the solution to the discrepancy to somehow lessen the effects of attributes.

I'm saying, there are two groups of people, and there's a discrepancy between them. If we kill one of them off, there's no more discrepancy.

As an aside, the same effect could be achieved by capping skills by attributes.

A correction, if I may. a 6 or 7 is not humanities' finest. It is unaugmented, mundane humanities' finest. It is John Henry, who magnificent as he was still lost to the steam engine.

I'm going to reassert that 7 is humanity's limit in 4th Edition Shadowrun. I'm pretty sure skills could exceed 7 in SR3, and they could just as easily go higher again in SR5. There is no reason to maintain a fluff that says 6 or 7 is a legendarily skilled character.





That's correct, because attributes are currently both too valuable and too easy to get. The value does need to be lessened. In the SR3, attributes applied only to the combat/spell/whatever pools, and general rolls were based only on skills. I don't necessarily want to go back to that, because the floating pools you can assign each pass/turn was pretty clunky and slowed the game down, but lessening the impact of attributes is something that has precedent and makes sense mechanically. What your attributes are should matter less than your training and skills.

Your solution is to instead make attributes more valuable than they already are, while encouraging people to have a high attribute to be highly skilled. I don't think that every runner in existence should have 6-9+ in every stat to be able to compete, and this is exactly what you are encouraging.

For Skills

Up to 6 cost normal, up to 9 costs x5, and over 9 costs x10.

Aptitude works normally, 7 costs normal, 10 costs x5 instead of 10.

It's true, that is what I'm encouraging, but that's probably worldview. If I'm not mistaken, you take a worldview where training is a bigger factor than natural capacity. I see them to be a little more equal than that, and I do think that skills are dependent on aptitude.

Skill having some dependency on aptitude/attribute isn't bad, but the current level of dependency is too high, and what you suggest makes it even higher.


Now, I don't like new skill costs scaling with each skill rank, so keeping with the idea of every 3 skill rating is a cost increase, but also a benefit increase, what about something like this?

-Skill costs 3 karma per level base. Every 3 levels in a skill past 1, your skill increase cost increases by 2.
-If your skill is higher than your attribute, increase the cost of increasing a skill by +2 for every 3 points higher the skill is than your attribute.
-Aptitude reduces skill cost by 1 per level, and lets you treat the associated attribute as 50% higher.


Examples:
-You have 6 in relevant attribute. Buying a skill from 1-3 costs 3 karma each. 4-6 costs 5 karma each. Since your agility is only 6, going from 7-9 costs 9 karma each, a very significant price jump.
-You have 8 in relevant attribute. Buying from 7-8 only costs you 7 karma each, but gaining your 9th point costs 9, since your attribute is now lower than the new skill rating. Going from 10 to 11 costs 11 karma, and gaining 12 costs 13 karma.
-You have 9 in the relevant attribute, and aptitude. Buying from 7-9 costs you 6 karma each. From 10-12 costs 8 karma. 13 costs 10 karma, and is the last point you get without increased cost from attribute (9*1.5=13.5=13), so 14-15 costs 12, 16 is 14, and 17-18 costs 16.


This method with uncapped skill progression does allow for much larger potential dice pools but generally only once you have several hundred karma under your belt. If you wanted to make base attribute more important rather than augmented attribute, then instead just make the karma cost for skills based on unaugmented attribute, and instantly getting past 6 skill becomes that much harder, and base attribute is far more important.

Your last two mechanics work well enough. I'd consider going with "Increasing the rating of a skill to anything above the rating of its linked attribute costs additional karma equal to the new rank minus the attribute's rank." It's a little harsher, but also simpler.

Also, it's probably simpler to just have Aptitude decrease karma costs.

I do have to ask, why don't you like the karma costs for skills increasing with each rank? Doesn't that perfectly match the idea of diminishing returns?

If anything, I think that the rate of increase isn't high enough, particularly in regards to learning time. Really, it should probably be exponential.

Hrm... that could work, I can see where tracking 2 separate intervals of 3 can be a little confusing. I could go for that if the attribute is augmented attribute. If you are instead using unaugmented that's really rough.




Simpler, but useless. Aptitude costs 20 karma to pick up. If you just make it decrease the karma cost, then it won't pay for itself until an absurdly high skill level. If you make it decrease skill cost and raise cap as I originally had it, then you get a little uncompensated karma, but you got the higher cap for that karma. If you have no cap, then you need to give something else, and giving an effective increase attribute for skill costs is the best way to emulate what was there before. It still won't be karma efficient for someone going to a level 9 skill, but someone aiming for 12 or higher will want aptitude to make it worthwhile.




Increasing with each rank leads to the skill getting too expensive too quickly. Unless you dropped it to like 1 karma per rank, which is too cheap. I figured 2 karma per 3 is a nice middle ground, and has the added benefit of making convenient breaking points. It also makes getting those early skill ranks a little easier. It's not so harsh to have a lot of skills around rating 3-4 when a rating 3 skill only costs 9 karma (currently 14 karma or 12 BP for the same deal). But then again keep in mind that I am in favor of average skill dicepools being increased in exchange for less bonus dice and stat contribution (the original calculation put getting 12 skill at about the same cost as getting to 7 skill now, and 9 dice around the same cost as 6 now).



Ugh. I'm exact opposite. Skill training time feels way too slow to me. Especially when initiation can be pretty much instant, increasing attributes is instant, etc. It may be a little less realistic, but multiple month training times really sucks in actual play. Especially when your GM doesn't like the idea of the group just taking a couple months off for downtime. (Earlier in the thread I complained about this as well as time for finding stuff)

Well, that's sort of the point. Ranks of 6 and 7 should be painful to get. I like the elegance that all skills and attributes work on a 1 through 6 scale the represents the range of human capacity. Even being able to attain a 7 as a human bothers me a little, but I'm willing to accept that it represents an abnormality in the species.

I just don't see the value in a larger scale. I do see value, though, in reducing the number of dice available from outside your Attribute+Skill.

We limit the number of extra dice to no more than your skill.

A simple and elegant solution.

The problem is you're still getting stuck on the 4th edition mechanics, while discussing potential changes going into 5th edition. 1-6 as the range of skill is what was determined arbitrarily for this edition. In the past, that was not the case. In the future it does not need to be the case either.

The value of having the larger scale of skill is to make a wider differential between someone who is really skilled and someone who is not. Right now the difference between someone who is average at a skill and someone who is legendary, is on average 1 success. The difference between someone untrained and someone legendary is just a hair over 2 successes. There just isn't really a lot of variability there. The fluff may say the gulf between these skill ranks is huge, but the mechanic doesn't support it.

With the 3 rank increase, every time you hit a threshold where you are on average getting 1 more success than an untrained joe, you are also making it harder to proceed further. The difference between a untrained guy and a legendary guy instead becomes around 4 successes. Someone can push themselves even further than that, but it becomes exceedingly expensive for relatively little benefit, so it's a easier pill to swallow.


Another alternative mentioned earlier is make the TN for a success 4 instead of 5, thus making smaller dice pool mods have a more profound effect. It also allows for a redesign of various thresholds to have some middle ground between "this is easy and anyone can do it" and "only a hyper specialist should be attempting this". But there's nothing really about that solution that makes it incompatible with any other changes.

I don't like it, it is in fact horrible IMO. Ignoring augmentations for the moment a scale of 1-7 means the best in the world is only 2 hits better than someone who is barely trained, assuming the same attribute. And with augmentations we are usually assuming the same attribute since it is so damn easy to cap it. So agile dude, who just learned how to shoot is basically just as good as the best marksman in the world, all because they decided to compress all abilities into a really small scale. It is a horrible mechanic.

In SR1-3 they allowed skills to go on for ever and the costs of increasing the skills eventually gets people to go to other things. As long as skills are priced right you don't need to artificially inflate them, since being tied to the rating increases the costs enough as is.