Linear advancement, Why not remove progressive costs as well Options

[Ellery]

I've been thinking about how best to dumb down SR and remove the need to think about what you're doing. I therefore suggest that SR4 will not have varying advancement costs or build costs associated with it. Everything will cost a fixed amount. After all, it's kind of complicated to figure out how much karma to spend on a skill at level 6. Wouldn't it be so much easier if it just cost, say, 5 karma per point?

Without a process of diminishing returns, though, people will probably spend all their karma on one thing to raise it to incredible levels. But there's an easy fix to that too--set a hard limit that nobody can exceed.

What do you do when you hit that limit? I can't hold 16 large dice in one hand comfortably, so let's say the limit is 10 for skills and 6 for attributes. When you reach 10 in your favorite skill, the path before you is simple! You spend karma on some other skill. Then, eventually, you've mastered too much stuff and it's time to retire because you're as good as it gets in everything important.

Would this be fun, or what?

[Critias]

What.

[sapphire_wyvern]

Cynical much Ellery? :)

Most of the "dumbing down" so far has been in mechanics invented wholesale by Dumpshockers. In fact, there have been cases of some unqualified people (eg me) being told by other unqualified people (eg you) that hypothetical mechanics are too complicated for the new allegedly dumbed-down system, based on no evidence whatsoever!

I suggest we all try to calm down. I think we're getting some positive feedback (in the bad, unstable behaviour-causing, control theory sense) going here.

[Hasagwan]

Boy I haven't seen that much sarcasm in a long time. I think we already have enough heated debates that we don't need a pointless one to start a flame war too.

While we all have our own opinions, at least everyone is trying to argue their points rationally and trying to make valid points. Maybe you should join us in that?

[Kagetenshi]

Wyrd. But I can hold sixteen dice, so I think the limit should be twenty, and you're a dumb-dumb for not thinking so too.

~J

[Critias]

It amuses me to see random DSers saying Ellery can't argue her points rationally.

[Hasagwan]

[sapphire_wyvern]

Well, I'm familiar with Ellery's previous posting, which has greatly impressed me with its excellent structure and logic, not to mention a commendable dedication to correct use of the English language.

However, in some recent examples (the discussion of whether or not spell Force will be incorporated in SR4) it is undeniably true that both of us having been arguing a case without evidence.

I argue that Force could be retained, possibly as some sort of cap on the number of successes that can be achieved on a spell. I have no evidence to support my contention that such a mechanic could be used.

Ellery says that that would be too complicated for a streamlined system. Ellery has no evidence to support that contention, either. I personally think that the idea of a cap on net successes on a check is a laughably simple proposition.

Ergo, Ellery (who is an unqualified person for the purposes of discussing SR4's mechanics) is dismissing my suggestion, which is obviously also made by an unqualified person. This is exactly what I referred to in my first post.

Making up mechanics, and dismissing them as too complicated, is no way to prove that SR4 is dumbed down.

I do not suggest that Ellery can't argue rationally. I would have to be particularly inobservant to do so. But in this case, Ellery's post is not a rational response to the information we have available for SR4.

As a joke, I think it's ok.

[Ellery]

There are some heavy doses of sarcasm and cynicism in my post, certainly.

But I wouldn't be posting it if I didn't think it was a serious possibility. A fixed cost to improve yourself is simpler. It works fine for improving attributes in D&D, and adepts can get more power points that way in SR3. Also, even SR3 uses fixed costs for the build system--what if they combine the build and advancement systems (for more simplicity) and just inherit the fixed costs there (except with higher limits so you have room to grow)?

So I'm asking: why not do it this way? Especially when you're busy building a new system and don't have time to worry about a character who has been played for five years.

"Because it isn't the SR I know and love!" isn't a credible answer, because plenty of people know and love combat pool--arguably more important than this--and the FAQ already dealt with that one.

"You're being sarcastic!" isn't a credible answer either, because being sarcastic doesn't prevent me from also saying something true.

"We don't know enough to say!" is quite true. We don't know enough to say for sure. But the problem is that by the time we know enough to say for sure, it will be the way it is, for better or for worse. Plus, in this case, there aren't many ways to do advancement. So here I think it's worth speculating and discussing. Maybe I should have started out with a little less sarcasm in the initial post. The goal here is to not start with SR3 and work downwards, but start with the simplest possible rules imaginable and work upwards until we hit something that might be good enough for SR4. That kind of gives us a range of possibilities, since presumably things aren't going to get more complicated than SR3.

In any case, if the discussions here are going to have any impact, they should be somewhat forward-looking, so if we say something insightful there's time for the devs to make changes accordingly. In this case, with so few options, I think we can be usefully forward-looking.

P.S. Thanks for the compliment, sapphire, and I agree on the magic thread that neither of us know what we're talking about. I wasn't even going to comment there, but maybe I was bored. I fully grant that spells could be learned at a given force, and that force could be used as a cap and to scale drain codes (assuming there is still drain), and so on. I think advancement is different, as described above.

[sapphire_wyvern]

Well, I myself have been advocating unified build & advancement rules. Unfortunately, by reading between the lines on what Synner has had to say on the topic, I'm pretty sure that the current version of SR4 does not have a unified b/a system. (Take that, NDA! ;))

I am more in favour of simply precalculating the skill and attribute costs from 1 to 6 to save time in chargen whilst retaining diminishing returns and escalating skill costs in future advancement. Best of both worlds to my mind, and stops players from maxing out skills and attributes without artificial restrictions (What, you're a skilled electronics expert? Well you can't be stealthy as well!)

In a sense, even a fixed advancement cost will have "diminishing returns" because your twentieth pool die will be a proportionally smaller improvement than your third.

But I am not in favour of "iron ceilings" in character development. It makes more sense to asymptotically approach perfection, rather than to linearly improve until the rules say "well, sorry, but you've now reached the maximum level of skill the universe permits".

You say it works fixed costs for improvement work fine in D&D. I say that this is arguable. For one thing, your improvement rates in different domains of competence (combat, saves, skills, spellcasting ability) in d20 is entirely defined by your class. Actual increases in attributes are doled out at a fixed rate, much as Karma Pool is in SR3. In a more flexible system, such an advancement paradigm is easy to abuse. Example: Decipher's CODA system used in the recent Lord of the Rings RPG, which is stupendously easy to rort.

In any case, the experience cost to level up in d20 (thus incrementing all your competencies by the class-defined amount) increases at each level; it's only because the experience rewards increase proportionally that D&D does not have diminishing returns itself.

[Ellery]

I also dislike iron ceilings, but I have noted, to my dismay, that game developers of all types love them. D&D has levels capped. SR is attribute capped. Star Wars: Galaxies has a skill cap. And so on.

They all irritate me, but they seem to be the industry standard. So I continue to wonder: why not flat, capped advancement? Or, alternatively, maybe they'll make any kind of advancement so expensive that there won't be much advancement.

If I got to pick how to do it, I'd make attributes exponentially expensive (keeping them more tightly in check) and keep the triangle progression for skills, providing tables for people who are turned off by anything that looks like (n+1)*n/2.

But my impression is that anything that could be viewed as mathematical is also viewed as complicated, so any ideas like that will be out. Personally, I think that roleplaying games are a great way to get people to want to improve their mathematical skills. Unfortunately, there's a pretty strong perception generally that math is scary and bad and geeky and will give you zits, so I'm pretty confident that any progression will have to be simple in the extreme. Also, I suspect that caps will be used extensively in place of diminishing returns even if a progressive system is used, since you can't tailor a simple progression to fit all cases without doing a little bit of math.

Removed in edit: a grammatically correct sentence that was, nonetheless, a paragraph long. Only philosophers have the attention span to read such things.

[Kagetenshi]

To some degree it already doesn't make sense that someone can start out with 1 Intelligence and end up at 9 (or even 3 and end up at 6 or 7). Likewise for Body, Strength, or Quickness; someone profoundly crippled enough to merit a one in any of those attributes is never going to hit Olympic level. I think capping attributes is a small price to pay for not blowing sense completely out of the equation.

~J

[Fortune]

I have no problem with capped Attributes. I just don't really want to see Skills capped.

[Ellery]

I'm sure you'll still be able to raise Int from Mike Tyson to Albert Einstein if they allow attribute advancement at all.

Exponents grow very fast. There's really never a reason to put a hard cap if you double the cost each level. Let's suppose an attribute of 6 costs 12 points. Then to get 7 would be 24, and 8 would be another 48, and then 96 for 9, and...look, you've got much, much better things to do with your karma at that point.

Added in edit: Though it would be an interesting mechanic to have the cost to increase an attribute depend on the number of times you raised it as well as where you started! For example, you could have it double each time after the first, but start at twice the current attribute. So if you start at 5, the next level would cost you 10, then 20, 40, 80. However, if you start at 3, it's 6 + 12 to get to 5, and then 24, 48, 96, 192 to keep going from there.

Hm, I really like that. It does mean that you have to work really hard to improve something innate that you're terrible at.

[RunnerPaul]

[Dawnshadow]

[blakkie]

I'll bite and assume this is serious and answer seriously:

The diminishing returns helps keep the mooks serious threats longer. If you change it to something close to linear power progression it screws with that aspect of SR3. Not that you could totally anyway unless you toss out the damage box system.

P.S. "Dumbing down" doesn't appear to be a Fanpro goal. Streamlining play is NOT dumbing down. Calculating skill and attribute costs is outside the normal play, which is where I think they are aiming at increasing the pace. Not that figuring out skill and attribute costs is all that hard for anyone functioning at atleast a 6th grade mathematics level.

[Ellery]

Maybe it isn't a stated goal, but it's easier to streamline by dumbing things down than by crafting a well-tuned set of rules that are understandable, resolve quickly, and nonetheless have a lot of rich complexity behind them.

The latter is hard.

Anyway, I think blakkie's argument is a decent one, but I'm not sure it's compelling enough to rule out flat advancement, especially since the TN5 thing will have already screwed with everything in SR. Since you need to rethink all of that (or have it break in not-so-amusing ways, such as an expert shooter shooting as well in the dead of night when he can't see a thing as an average shooter does on a bright sunny day). So I don't think they're being shy about screwing with aspects of SR3.

[nezumi]

I like diminishing returns. Most people should be 3 or below in any given skill. 6 really should be special. If it's just as easy to get to a 6 in something as a 3, it steals some of the thunder, and it'll lead to serious inflation in peoples ratings all around.

Just because you got an associates doesn't mean you should be able to get a doctorate with just as little work, spread over more time.

[Spookymonster]

Not sure if it's been mentioned, but what about fixed costs for skills, and exponential costs for attributes, e.g., each skill level costs 6 kp, but each attribute level costs (lvl * lvl) kp?.

This would gradually make it cheaper to buy skills than boost attributes - for example, raising Int from 5 to 6 (36kp) vs. raising Pistols from 3 to 5 (12kp), SMGs from 4 to 6 (12kp) and Rifles from 5 to 7 (12kp).

[Ellery]

I agree, nezumi, except I am not fixated on any particular number. Some skill should be average, and some higher skill should be really exceptional, but exactly what those values should be depend on the dice and skill systems, unless you want games where the best specialists in the world are not much better than average people with a little training (or where one combat specialist can take on a thousand non-specialists at once and win--but I don't see any danger of that with the currently announced changes).

If an average skill is 3, and a good attribute is 5, then someone with natural talent and average skill will get 8 dice. Someone with rare, specialized skill will get only three more dice--11 dice vs. 8 dice is not that big a deal. If you zoom into the head-to-head probability chart on p.11 of the dice analysis paper I linked to, you'll see that at 11 dice vs. 8, you only win 75%-80% of the time if you win ties, and about 60% of the time if you lose ties.

That's not much of an advantage for supposedly being really good in comparison to the person with average skill.

[Vuron]

I'd have to say dumbing down definitely doesn't seem to be a design goal but improving flow of play and interaction between archetypes at all points in the game does seem to be a definite focus.

So I want Strength 5 (Natural) that costs 20 points and Edged Weapons 4 that costs 8 and Lightning Reflexes 2 costs 10 points etc. Yeah it's a doable system and I've seen it used to decent effect in settings where basically everyone is the same archetype but in general going with fixed costs tends to eat up huge page counts from books. Formulas for advancement can often take up 2 pages and still have room for charts and examples.

However I do think you have made a good point in one of you later arguments as to exponential advancement reducing the needs for hard caps. While I still tend to think that attributes will be purchased to the racial maximum (lets say 6 for humans) with a simple formula like x^2. After 6s you switch to a formula of doubling the karma cost for each additional level.

Human Attributes Karma Cost

Level 1 1
Level 2 4
Level 3 9
Level 4 16
Level 5 25
Level 6 36
Level 7 72
Level 8 144
Level 9 288
etc.

That way you don't have to have maximums at all. Given enough Karma and no common sense you can have Mundanes with Strength 20 that can wrestle with Kodiaks and Juggernauts. Of course anyone with half a brain will settle for something like strength 7 or 8 unless they really want to resemble hercules ;)

[blakkie]

Do you realize Vuron that (Level + (Level-1)+...+1)*2 is the same value as (Level+1)^2? Think of (Level + (Level-1)+...+1) as the area of a triangle (which is basically what it is). Two triangles is a square (in the case of these special triangles). So up till 6 your proposal is significantly CHEAPER than than in SR3. Going from 6 to 7 does become slightly more expensive, but still going from 3 to 7 using your progression is still a lot cheaper in Karma overall.

[Vuron]

[blakkie]

You do have to remember to keep it in balance with skills cost. Unless they ramp up the number of [useful] skills significantly, if you make the attribute too cheap then skills become less important as boosting the attribute boosts die across all linked skills.