I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.
Accepted, but can you really not see why someone else wouldn't have fun?
That's very close-minded IMO. The player has been playing fighters (he probably means martial actually) and wants to play a full caster. All of these classes accomplish this.
You say close-minded, I say having a clear vision of what I want to play. Are you saying that that is badwrongfun?
...but because point buy is bad for PF1..
If it was that bad it wouldn't be so popular. If it was objectively bad why do so many people never ever want to go back to rolling. I mean I've rolled for stats far longer than I've used point buy. I would not go back voluntarily.
And if you are using a different method for rolling that grants you higher expected results for secondary and tertiary stats then you can achieve the same result by increasing the PB value.
The issues with game design are not changed by your stat generation method. Your stat generation method will not fix any issues with game design.
Again, you're basically just reiterating my point.
Yes, I use a different rolling method, just like the entirety of PFS players use 20 points instead of 15.
Yes, the problem is with asymmetrical game design.
Point buy facilitating your needs is not my problem, claiming that it is in any way balanced is. Some people thoroughly enjoy using point buy as a parameter for engaging their resource management skills as part of character creation. Some don't like having those resources restricted before they even play the game. I use rolling because it doesn't necessarily guarantee that limitation, but I've always been open to better stat methods. 3d6 replace lowest with 6 was my solution, and I've found that my players are relieved that they don't feel shoehorned into a specific finite curve of options before they even write down their stats. In table experience, it also means the big six become less important when players have decent enough stats on their own to avoid the need. (For example, a cleric that can get higher WIS can get an extra spell slot that casts Shield of Faith that scales better than buying a ring of protection, and can even afford two pearls of power to have even more of them effectively nixing the need for that item as part of the character's investment.) This was a positive change at my table, it might be negative at yours if it doesn't match your style. I'm resistant to use any metaphor involving a box, but point buy servers as that metaphorical box. You're forced to function within that trigonomic relationship between stats, which seems balanced only because everyone has to build their characters within that same mathematically designed box. Funny thing about boxes, they're inefficient compared to spheres as far as individual capacity goes, though they tessalate well which makes them seem superior when you have an environment whose model is based on organized play.
Then there's the issue with the game needing patches to fix this problem because mainly the organized play couldn't accommodate this asymmetrical design in the game.
Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
This sort of analysis made me curious how the method I use as a DM stacks up, I don't tend to get much in the way of complaints and so I was curious.
I created a spreadsheet that mimics my method with the click of a button and with the results I got It's not surprising.
The effective point buy value of the highest array ranged from 29 to 41, with most falling around 37 to 39. This also explains to me why even a 25 Points never feels like enough and why balance is rarely an issue, if everyone is around 38 PB, players are free to pursue whatever build they want to without feeling weaker than the next person.
That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us less lucky players would end up with a 15pt buy build. It is certainly possible to have fun with either 45pts or 15pts--I don't think many folks claim otherwise although some are less enthusiastic about working with only 15pts. The point being made, the one that you have steadfastly overlooked, is that it is significantly more difficult to have fun when some of the characters in a group have 45pt buy builds and some have 15pt buy builds compared to when all characters have equal point buys to work with. It is also significantly more difficult for the GM to challenge a whole party like that because he's either boring the high point buy builds with challenges equal to the low point buys or he's slaughtering the low point buy builds with challenges equal to the high point buys.
That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us less lucky players would end up with a 15pt buy build. It is...
Your premise assumes that two builds using the same amount of points will be equally effective. This is verifiably false, with math.
PBEs are a useless metric, unless two players are making comparable builds, this is because the classes, and their attribute dependency, is/are asymmetrical.
Point Buy is not balanced. It's a lie to say that it is. The classes don't even start out with the same amount of starting gold in the core rules, the classes are not balanced.
Now, in the 15pt vs 45pt scenario, you can definitely have two characters on the same team who don't even notice the disparity, because of the asymmetrical needs of the classes. If players feel personally disserviced by this, well that's a problem for the individual player.
This reason is why I let my players trade stat arrays if they roll for a better wizard when they wanted to play a paladin, et all.
| 1 person marked this as a favorite. |
...That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us less lucky players would end up
Your math means very little when actual our play experience is to the contrary, just like the folks who insist in combat healing is always a waste of time when our play experience is that we have avoided TPW's only due to in combat healing on numerous occasions.
We noticed a stark difference in the performance of the characters when we rolled for stats. The GM also noticed it was difficult to safely challenge the party when we rolled for stats. It's not like the GM woke up one day and decided we would use point buy for no reason; we had problems we were trying to solve, we noted a massive disparity in stats, made the change to point buy and discovered that these problems were both mitigated as a result.
It is possible that we could have used some formula to calculate the optimal point buy for each type of character being rolled so that a SAD class is not unfairly advantaged by point buy compared to a MAD class with that same point buy. It is also possible that we would be even happier if we did such a thing but that's an awful lot of work compared to just switching everyone to an equal point buy, especially when most of us have done little beyond basic mathematics in 20+years.
I've never claimed that the classes are balanced or that point buy creates balance. We do feel that simply decreasing the amplitude between the peaks and troughs so everyone's stats are closer than they were when we rolled has been enough without achieving perfect mathematical balance.
...That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest
That's fair, as I haven't played at your table so I cannot attest to your actual experience.
In mine, point buy stagnated character creation, and the mathematical flaws proved to be more trouble than using heroic rolling methods, since that is meant to be the equivalent of the 20 point buy anyway.
You actually get a lower amplitude of variance with 2d6+6 than you get with 20 point buy. My group(s) over the years have had varying degrees of mathematical inclination, with the group I started PFRPG with after switching from 3.5 dubbed it "Marshall Math" and the DM gave up on auditing me because I was literally never wrong. They were a much more casual group than I'm used to playing with, however.
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.
I'd still find myself at the bottom of that bell curve - dice hate me...
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.
math is fun for me so I ran a simulation of your method and it seems like you may be low-balling.
If I assume that a player only gets a single array the Equivalent Point Buy numbers are as follows.
Low: 16
Median: 31
High: 49
Average: 31.8
Using your method and allowing players to pick from 4 arrays rolled as such
Low: 39
Median: 39
High: 49
Average: 43.5
Unless, I've missed something with point buy costs.
I'd still find myself at the bottom of that bell curve - dice hate me...
Me too. That's why I use such a generous method.
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.math is fun for me so I ran a simulation of your method and it seems like you may be low-balling.
If I assume that a player only gets a single array the Equivalent Point Buy numbers are as follows.
Low: 16
Median: 31
High: 49
Average: 31.8Using your method and allowing players to pick from 4 arrays rolled as such
Low: 39
Median: 39
High: 49
Average: 43.5Unless, I've missed something with point buy costs.
I'd still find myself at the bottom of that bell curve - dice hate me...Me too. That's why I use such a generous method.
An individual having a mean of 31 is not that far off from my estimate.
This of course, doesn't take into the effect that some classes are more like spheres, some are more like squares, but point buy forces us to treat them all like triangles.
Funny thing is, if you end up at the bottom of the curve, you'll most likely end up on a chassis that matches a SAD character and in my system you can trade with the player who wants to play a SAD class.
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.
That's the 2d6+6 method which I think is...decent. You can still get flat arrays out of it and a modest stat disparity which are its only weak points. On average just shy of half of the character's stats will be in the range of 12 and 14.
If I were using this method, I'd let players re-roll if they got two stats below 10 or no score above a 14. That's going to happen maybe 15% of the time?
(edit to fix that percentage)
For example, character #3 would not be much fun. I'd be willing to play character #1 but I'd be irked if most people looked like #2.
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.That's the 2d6+6 method which I think is...decent. You can still get flat arrays out of it and a modest stat disparity which are its only weak points. On average just shy of half of the character's stats will be in the range of 12 and 14.
If I were using this method, I'd let players re-roll if they got two stats below 10 or no score above a 14. That's going to happen maybe 10% of the time?
2d6+6
2d6+6
2d6+6
2d6+6
2d6+6
2d6+62d6+6
2d6+6
2d6+6
2d6+6
2d6+6
2d6+62d6+6
2d6+6
2d6+6
2d6+6
2d6+6
2d6+6For example, character #3 would not be much fun. I'd be willing to play character #1 but I'd be irked if most people looked like #2.
It's one half-step up form 2d6+6, since you get to still drop a die. I wanted to find a middle between 4d6 drop lowest and 2d6+6.
Slightly higher averages, and the players don't feel gipped.
EDIT: also, does no one use the rule of 72? Your stats totaled up must equal or exceed 72 or they get thrown out. From my players telling me this, it was the norm even in 3.5.
I'd recommend trying out 3d6, treat lowest as 6 for anyone looking at rolling methods who want to shorten the width of the bell curve. It should roughly equal 25 point buy on average, assuming I did my math right.That's the 2d6+6 method which I think is...decent.
It’s not the same distribution. The easiest way to see the difference is to consider the likelihood of rolling an eight:
Your method will produce a score of eight 1/36.
3d6 replace the lowest with a six gives an eight 1/216.
It's one half-step up form 2d6+6, since you get to still drop a die. I wanted to find a middle between 4d6 drop lowest and 2d6+6.
Ah. I misunderstood. My apologies!
I like it better. Rule of 72 doesn't mean much to me. A flat array of all 12's passes.
At minimum, I like to see one high stat (>=15), one almost-as-high stat (0 to 3 points below the high stat) one low-ish stat (8 to 10), and the rest in the range from 10 to the 2nd highest stat. That feels like a character that has a weak point, a strong point, an almost-as-strong point (to allow for MAD characters), and no other penalties.
But I am also a weirdo who loves automating complicated algorithms.
That's very close-minded IMO. The player has been playing fighters (he probably means martial actually) and wants to play a full caster. All of these classes accomplish this.You say close-minded, I say having a clear vision of what I want to play. Are you saying that that is badwrongfun?
I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
At minimum, I like to see one high stat (>=15), one almost-as-high stat (0 to 3 points below the high stat) one low-ish stat (8 to 10), and the rest in the range from 10 to the 2nd highest stat. That feels like a character that has a weak point, a strong point, an almost-as-strong point (to allow for MAD characters), and no other penalties.
But I am also a weirdo who loves automating complicated algorithms.
I wonder if there’s a method where the dice change based on how lucky you’ve been. Something like:
First roll = 3d6
Second roll = (4d6 unless the first roll was 16+ in which case it’s 3d6)
.
.
Roll x = (2+x-y)d6 where y is the number of previous scores 16+
You could weight multiple 16s more heavily (so maybe y=4 if you’ve rolled three 16s)
At minimum, I like to see one high stat (>=15), one almost-as-high stat (0 to 3 points below the high stat) one low-ish stat (8 to 10), and the rest in the range from 10 to the 2nd highest stat. That feels like a character that has a weak point, a strong point, an almost-as-strong point (to allow for MAD characters), and no other penalties.
But I am also a weirdo who loves automating complicated algorithms.
I wonder if there’s a method where the dice change based on how lucky you’ve been. Something like:
First roll = 3d6
Second roll = (4d6 unless the first roll was 16+ in which case it’s 3d6)
.
.
Roll x = (2+x-y)d6 where y is the number of previous scores 16+
There's the dice pool mechanic, where you get a set number of dice, and allocate them based on what stat you roll for, with mins and maxes based on the size of your pool.
The method I was suggesting would kind of “correct” for good/bad luck and would all but guarantee the 16 John is looking for without building in high stats all over the place.
I wonder if there’s a method where the dice change based on how lucky you’ve been. Something like:
As mm says, the dice pool methods do this pretty well.
When I was writing my little stat array generator, the approach I took was to simply bound each score differently, using its own baseline, die type, and drop method (drop high or drop low). Then I restricted the final ranges to a narrow pt-buy range because it was convenient and that's what people tend to think in.
It artificially implements what you describe: the top tiers are "luckier" than the lower ones, and if you get too lucky (or unlucky) it silently re-rolls everything.
I wonder if there’s a method where the dice change based on how lucky you’ve been. Something like:As mm says, the dice pool methods do this pretty well.
The difference I’m looking for is trying to generate the PC weakness you wanted as well as to minimise differences between the PCs. A dynamic system will autocorrect to reduce the disparity you (and others) object to.
Dice pool methods often produce even worse differences (the player rolling a fistful of one’s with their 9d6 stat is not only unlikely to do well overall - they’re also stuck with a subpar key ability score).
Computerising things and rerolling outside of certain bounds is the technically best solution, I think.
It just feels like cheating! ;p
| 1 person marked this as a favorite. |
Computerising things and rerolling outside of certain bounds is the technically best solution, I think.
It just feels like cheating! ;p
I've done my method by hand and it's too much like work. :) So, yeah, I don't suggest it for folks who like the tactile nature of physically rolling dice for their scores.
That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us less lucky players would end up with a 15pt buy build. It is...
Hello, I don't agree with what you say here that I am answering different questions than I am being asked. IMO I am answering the question than talking about something related. For example, here I start off by saying that I would play the character alongside a 45pbe (which really doesn't mean much IMO) then going on a rant about how I can have fun like that. I thought it was clear that I was talking about playing alongside the other character especially when I explicitly talked about the other character and my character's relationship.
I was talking about us being in the same party and I think that we can have fun together because that is what the game is about. And yes, the GM may have a hard time making challenges for the party but as long as there is no Int/Wis damage/drain I think my Swashbuckler (let's go with that) can contribute stuff to the party and I could have fun doing it. Some GM tactics (that I can think of off the top of my head) are walls of [insert element here] in between the "Mary Sue" and the rest of the party, challenges from bosses while others fight minions, straight up targeting by monsters (you can invent excuses as to why), making monsters that are immune to some of the "Mary Sue's" tactics or expose his weaknesses, etc
| 1 person marked this as a favorite. |
Character weaknesses are imposed more by classes than stats ever will, at least in PF1.
Decent stats are the only defense to mitigate this in practice, along with magic items.
...That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us less lucky players would end up
My apologies, I didn’t realize the friend you were referring to would be whatever character was made using the 45pt buy from the example.
That said, “as long as there is no Int/Wis damage/drain I think I can contribute” is admitting to a pretty glaring weakness that your 45pt buy friend will likely never have to concern himself with. What will your swashbuckler contribute aside from stimulating roleplay? Are you going to stand in melee rocking that 10 con? You have 1 skill point/level. Your perception will always be atrocious. OTOH, your 45pt friend has a 4 point advantage on int related checks, at least 2 more skill points per level, a 3 point advantage on wis related checks, at least 2 HP more per level and his important saves are vastly superior to yours. While you can choose feats like Improved Will, Great Fortitude, Cunning, Toughness etc. to help bring yourself on par with him, he has the options to choose those as well. Or he could take feats that actually improve his character rather than making it just barely competitive. These things are meaningful at my table so we stopped rolling for stats.
Certainly you can play anything but in my experience, encounters that challenge you will generally be a boring walk in the park for your 45pt friend and encounters that challenge him will generally destroy you. Unless, of course, your GM takes pains to make the 14pt buy character viable alongside the 45pt buy character. Which my GM never cared to and I don’t blame him considering the hours he puts into the game already without us players adding even more for no particular reason or benefit.
| 1 person marked this as a favorite. |
...That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt buy build while the rest of us
I need the actual information on the builds these different PBEs are piloting to compare them.
A wizard on a 15 PBE is going to be nigh exactly as effective as one on a 25 PBE etc.
Are you comparing two fighters? A fighter vs a wizard?
PBEs are a useless metric unless we're comparing the same class, or classes with the same level of MAD.
Different classes already offer differences in HP/level. They already offer disparity in rates of saving throws. They already have asymmetrical numbers of skill ranks per level.
The classes are not designed to be homogeneous, so comparing PBEs with no context offers no real information other than you value PBEs more than you value anything else.
MATH
He’s talking about some kind of martial alongside a swashbuckler. He said fighter but I’m not sure if he intended the class. And we’re not talking about 15pt vs. 25pt. There’s not much difference between those two. 45pt and 14pt OTOH, well, I just detailed what some of the differences are and they are significant regardless of the class. Surely with all the MATH you refer to you understand the effect of disparate baselines.
| 1 person marked this as a favorite. |
That's very close-minded IMO. The player has been playing fighters (he probably means martial actually) and wants to play a full caster. All of these classes accomplish this.You say close-minded, I say having a clear vision of what I want to play. Are you saying that that is badwrongfun?I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
And when Bob sat down to play, he had a wizard in mind. Now he can't play one.
If Bob sat down at your table, rolled those arrays, then said "but I really wanted to play a wizard", would you argue with him that he's doing it wrong? Because that's a great way to tell Bob that you don't want him at your table.
That's very close-minded IMO. The player has been playing fighters (he probably means martial actually) and wants to play a full caster. All of these classes accomplish this.You say close-minded, I say having a clear vision of what I want to play. Are you saying that that is badwrongfun?I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
And when Bob sat down to play, he had a wizard in mind. Now he can't play one.
If Bob sat down at your table, rolled those arrays, then said "but I really wanted to play a wizard", would you argue with him that he's doing it wrong? Because that's a great way to tell Bob that you don't want him at your table.
The conversation about rolling vs pointbuy should happen way before deciding to play, in my view.
If you go into a game where you’re going to roll for stats, you’d be foolish to approach it from the point of view of “trying” to roll stats to suit a particular character concept.
Irrespective of what one’s preferences are, you shouldn’t agree to play a game and then be disappointed when the rules you agreed to play out as expected. (I’m presuming the situation where everyone is reasonably au fait with the pros and cons of this choice).
Agreeing to roll for stats is agreeing to give up some agency in creating your PC.
I think both sides have valid points, we are just looking at extreme examples because two of the rolls created that and it is also easier to make a point when that happens.
I agree that class disparity tends to be a bigger issue than attributes. I play a wizard with a Str of 7 (4d6 drop lowest, with one full regen due to lousy stats). She is extremely effective as long as she stays far, far away from melee.
But the mechanical impact of that 7 on her as a character is not zero. She almost stumbled after the party into a den of shadows. Had I not hesitated, she would not have come back out. Our martials took more Str drain than she has Str.
I am not complaining. I love this character and the rest of her stats are great. But the resources expended to boost that are resources I don't have for other things.
He’s talking about some kind of martial alongside a swashbuckler. He said fighter but I’m not sure if he intended the class. And we’re not talking about 15pt vs. 25pt. There’s not much difference between those two. 45pt and 14pt OTOH, well, I just detailed what some of the differences are and they are significant regardless of the class. Surely with all the MATH you refer to you understand the effect of disparate baselines.
At low levels, sure, the math supports that claim.
At higher levels when flat bonuses from class progression and magic items supersede the initial attribute allocation you hardly notice a difference.
The swashbuckler is specifically designed to be able to be built as a two to three attribute class, since it uses CHA to qualify for combat feats in place of INT, and was designed in the same book that Core DEX to damage was intended to be available. DEX/CHA/CON means aside from making sure you don't overload your carrying capacity, you're pretty much set as far as attributes go with the same kinda build you could see on a sorcerer.
Comparing it to a vanilla fighter, he doesn't even get to use both Armor Training and Weapon Training [effectively] unless it's specifically a DEX build itself. I shan't get into the nuance of tricks the fighter has to mitigate its problems with the design (all of them were patched eventually) but the fact still holds that the newer classes like the swashbuckler are intentionally designed around the paradigm of point buy.
Just gonna pipe up and say I've never heard of the "rule of 72" before today.
That's very close-minded IMO. The player has been playing fighters (he probably means martial actually) and wants to play a full caster. All of these classes accomplish this.You say close-minded, I say having a clear vision of what I want to play. Are you saying that that is badwrongfun?I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
And when Bob sat down to play, he had a wizard in mind. Now he can't play one.
If Bob sat down at your table, rolled those arrays, then said "but I really wanted to play a wizard", would you argue with him that he's doing it wrong? Because that's a great way to tell Bob that you don't want him at your table.
No, I would say, "Hey Bob, I know you really want to play a Wizard but I don't think you rolled the stats for that. How about you play a Sorceror instead, their like Wizards but they don't have to prepare their spells! I think you'll have a lot of fun."
| 1 person marked this as a favorite. |
...That's the easiest question I've heard all day. Yes, yes I would. I am very confident in my ability to have fun playing my dumb but good natured Sorceror struggling to control his powers, or maybe my naive Swashbuckler trying to understand the world and make his friends proud along with my friend the strong and reliable fighter. Maybe he is my older brother/sister looking out for me, or maybe I'm a charming idiot they met on their travels. I've seen people have fun with 3 Str characters and I believe I can have fun with a 4 Int character.Roll 4d6 drop lowest in order.
Add/retract racial bonuses.
Work your characters around overall results of random rolls and characters description given earlier.
So let's try that.
[dice=Str]4d6-2
[dice=Dex]4d6-4
[dice=Con]4d6-2
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Yowza! I don't know anyone who would turn down a 45-pt build.
How about character #2?
[dice=Str]4d6-1
[dice=Dex]4d6-3
[dice=Con]4d6-1
[dice=Int]4d6-1
[dice=Wis]4d6-1
[dice=Cha]4d6-1Uhhhh....That's technically a 14 pt. build. But more importantly, would you want to be playing character #2 in the same campaign with character #1?
This is the problem with flat-formula die rolls.
You are repeatedly answering a different question than you are being asked. You keep telling us that you could have fun with a character using the second set of stats. No one has asked you whether you could have fun with that set. The question is how much you would want to play a character with those stats alongside another character using the 45pt build.
My group switched point-buy after many, many years of rolling stats because we have one particular player who would reliably roll a 45pt
The quotes lost them but the builds we're something like 18,16,17,14,15,14 and then 11,16,10,4,6,16 I think. I saw the stats for the second one and thought Swashbuckler or Sorceror but I started using Swashbuckler as my example so that kind of stuck. Now that I'm actually thinking about it Sorceror would be better because Sorceror is SAD and doesn't care about that Con being 10.
| 1 person marked this as a favorite. |
No, I would say, "Hey Bob, I know you really want to play a Wizard but I don't think you rolled the stats for that. How about you play a Sorceror instead, their like Wizards but they don't have to prepare their spells! I think you'll have a lot of fun."
It'll never happen, because as the Marshmallow says you'd have had the chat about rolling/PB before hand in an ideal world.
This however is not the answer to someone who dislikes rolling. The reason I prefer point buy is because I want to play the character I envision in my head - unbeholden to dice - If I wanted to play a sorceror or a kineticist then that is what I will build. If I wanted to submit to the fate of the RNGods and see what I get then I would be happy rolling. But I don't, I have a vision in my head of a particular character who has a high intelligence and memorises spells - I want to play a wizard, not something that is not a wizard. That is my fun.
Ultimately, your idea of fun turns me off and has for many years. I have created far more characters decided by the dice than not - it is not as much fun for me. I prefer to determine my own fate. You telling me "you'll have fun playing not-a-wizard" is quite frankly b/s. I might have fun, but not as much fun as if I get to play the character I want to play.
That is what (eeps!) 30+ years of gaming has taught me. There are people who have been gaming longer than I have and they still get the thrill of rolling and rolling with whatever that throws at them - good for them. It is however not for me. From the very first character I built with PB I was sold. No more suffering at the vagaries of the dice, no more compromising except where I chose to compromise. That is my fun - Sure I'll still roll if I have to, but I'd much prefer not too.
So I guess the real answer to anyone asking is "try both and see which works best for you and your group". It is the only proper answer to very many of these types of questions.
| 1 person marked this as a favorite. |
I'm a fan of 15 pb. Too many rolling players cheating their rolls. Too many bad memories. Plus, every roll-preferring player I've ever met, when asked why they want to roll, have given answers indicating that it's because they want their character to be better than their allies' right from the start.
15 pb keeps people more even, divided only by skill in assigning their stats, and I'm happy to help newer players. I've heard it encourages minimaxing, but my experience has been the opposite. When players need to dump everything hard to get the 20s they want, they tend to shift their goals to more reasonable numbers and buy more 14s and 16s.
...And finally, 15 pb let's me play more colourful opposition. I can run kooky builds and unusual tactics against the PCs, and not worry so much about needing to optimize the enemies to keep up with them.
|
| 1 person marked this as a favorite. |
Currently using 4d6 reroll 1s drop lowest, with rerolls allowed for particularly low arrays essentially by GM fiat.
It's... worked so far but I'm looking for a new method. Priorities in choosing a stat method for me are:
1) Allows people to create the character concept they envisioned. Most of the players in my group have a pretty good idea of what they want to play when they start to make their characters, and anything that makes it impossible to play eg an elven monk because you rolled badly is a non-starter.
2) Feels organic. Point buy just doesn't feel as interesting to me - ability score generation is one area of the game where I prefer a bit of surprise to another resource management exercise.
3) Balanced/fair. Ability score arrays certainly aren't the only source of mechanical imbalance between characters in a game, and we're pretty good at making everyone feel equally relevant even if there's a bit of a range in stat arrays. But it's definitely preferable if some players aren't noticably stronger or weaker because of simple luck during character creation.
I'm strongly tempted to try the "everyone rolls an array and then each player picks one of these arrays" method for the next game I run. It would be an improvement on fairness and within-party balance, be just as organic as everyone rolling their own array, and if anything increase the odds that you'll be able to find an array in the pool that fits your character concept (even if it's not necessarily the highest point buy array in the pool).
I'm also really intrigued by "organic" methods involving rolling in order or assigning dice to stats and then rolling, and have started using these methods to create NPCs with unusual talents or weaknesses. But I don't think it could ever be my group's primary method for ability score generation due to (1). Maybe for a one-shot, or if it was offered as an option alongside a more flexible method (possibly with slightly more d6s involved to make up for the likely non-optimal stat results)?
|
16 15 14 12 11 10 pre-racial array. 25-point buy equivalent. Has a couple odd stats so that it makes your stat bumps at 4 and 8 feel like they're worth more than nothing at 4 and a +1 to your main stat at 8. Its a sweet spot and still gives a bit of play-room for your racial -2's to not turn a stat into a dump stat.
|
|
No, I would say, "Hey Bob, I know you really want to play a Wizard but I don't think you rolled the stats for that. How about you play a Sorceror instead, their like Wizards but they don't have to prepare their spells! I think you'll have a lot of fun."I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
And when Bob sat down to play, he had a wizard in mind. Now he can't play one.
If Bob sat down at your table, rolled those arrays, then said "but I really wanted to play a wizard", would you argue with him that he's doing it wrong? Because that's a great way to tell Bob that you don't want him at your table.
If you genuinely don't understand why Bob would be disappointed by that, I will try to explain.
When I come up with a character concept, that concept is often rather specific in ways that indicate either a single class or a very short list of classes, and have some sense of the character's key ability scores built in.
For example, I don't want to play "a spellcaster with a reverence for nature." I want to play a wise, grey-haired mystic with a talent for shapeshifting - a druid or maybe a feral hunter, not a nature oracle or sylvan sorcerer or even a cleric with the animal and plant domains.
I don't just want to play an arcane caster. I want to play a bookish academic mage obsessed with planning for every eventuality. Having a high Intelligence score, good Knowledge skills, and even prepared spellcasting are part of the concept. I could still have fun playing a sorcerer, but it wouldn't be the same character concept and I likely wouldn't be as invested in it as the character that I'd already started to develop before stats were rolled.
Even if I don't have a class in mind, I might have a broad but stat-dependent sense of what I want to do like "agile fencer" or "guy with a really big axe" or "charming con-artist" which don't work if I end up with a low dex, strength, or charisma, respectively.
No, I would say, "Hey Bob, I know you really want to play a Wizard but I don't think you rolled the stats for that. How about you play a Sorceror instead, their like Wizards but they don't have to prepare their spells! I think you'll have a lot of fun."I genuinely don't understand what you're saying. I literally said there are more ways to have fun, how is that saying you play badwrongfun. I'm trying to open up more options not say existing ones are badwrongfun.
Question: Why do you have to play a Wizard? That actually doesn't make sense to me. Class in Pathfinder is just a means to an end, something you wrote on your character sheet. You can play the same concept and do the same things in many different classes.
Personally, when I create a character I think of what I want them to be. For example, for one character I wanted to play a daring pious catfolk martial. I considered multiple classes such as fighter, slayer, rogue, warpriest, and prestige classes such as evangilist, shadowdancer, and defensive stalwart (or whatever it's called) before deciding on swashbuckler/shadowdancer. The class isn't the first choice for me and I don't see why it has to be for you.
And when Bob sat down to play, he had a wizard in mind. Now he can't play one.
If Bob sat down at your table, rolled those arrays, then said "but I really wanted to play a wizard", would you argue with him that he's doing it wrong? Because that's a great way to tell Bob that you don't want him at your table.
If you genuinely don't understand why Bob would be disappointed by that, I will try to explain.
When I come up with a character concept, that concept is often rather specific in ways that indicate either a single class or a very short list of classes, and have some sense of the character's key ability scores built in.
For example, I don't want to play "a spellcaster with a reverence for nature." I want to play a wise, grey-haired...
I guess this makes sense but me and my group don't play like that. Anyway, the way I roll people still get to assign their ability scores in any way they want so this doesn't actually happen.
Moving on, I just came up with this a little while ago (though someone probably already thought of it) is to roll 1d6 6 times, 1 for each stat but you add it to a base of 2 6s, 2 10s, and 2 12s. Then you assign them to any score. This gives you multiple good stats, multiple ok stats, and multiple bad stats while mantaining a range of 7-18 and making sure you have the stats you want in each ability score.
I'm going to test it here (I hope this works)
Wow those are low rolls but you see what I mean.
| 1 person marked this as a favorite. |
I wonder if you’d do better with adding 2d6? (With bases of 3, 6 and 9 (drop anything over 18)).
The more dice you roll, the more concentrated the results will be around the average (hence the less disparity there’ll be between multiple PCs). That seems to me to be the most widely held objection to rolling.
If you do anything to limit the range of rolls to reduce disparity, or limit the impact of bad rolls I have to ask why you don't just find a satisfactory level of PB? Surely there comes a point where the difference between rolling a single d6 to determine stats doesn't really create any significant variance?
I mean that was one of the things that turned me off rolling, was the convoluted ways to ensure you had reasonable array (in addition to pointless re-rolling until you got an array you could live with, and the dissatisfaction with having to proceed with an unsatisfactory array).
Not making a point but this does remind me of shenanigans I have had to deal with: I used to play with a guy who tried to get away with rolling the dice one at a time and if it came up low he would aim the next dice at it to try and change it.
If you do anything to limit the range of rolls to reduce disparity, or limit the impact of bad rolls I have to ask why you don't just find a satisfactory level of PB? Surely there comes a point where the difference between rolling a single d6 to determine stats doesn't really create any significant variance?
I mean that was one of the things that turned me off rolling, was the convoluted ways to ensure you had reasonable array (in addition to pointless re-rolling until you got an array you could live with, and the dissatisfaction with having to proceed with an unsatisfactory array).
Not making a point but this does remind me of shenanigans I have had to deal with: I used to play with a guy who tried to get away with rolling the dice one at a time and if it came up low he would aim the next dice at it to try and change it.
That’s pretty lame.
I agree with you on the skepticism re:convoluted rolling constraints. I guess it might be more about feel than results. You still get the fun of the unknown without the risks of what are perceived to be the downsides.
Personally, the adjustment I’m comfortable with is “anyone can choose pointbuy instead of rolling”. Other than that, the potential of rolling a “sidekick character” is part of the attraction to me.
Neurophage, I love your approach to things
If you do anything to limit the range of rolls to reduce disparity, or limit the impact of bad rolls I have to ask why you don't just find a satisfactory level of PB? Surely there comes a point where the difference between rolling a single d6 to determine stats doesn't really create any significant variance?
I mean that was one of the things that turned me off rolling, was the convoluted ways to ensure you had reasonable array (in addition to pointless re-rolling until you got an array you could live with, and the dissatisfaction with having to proceed with an unsatisfactory array).
Not making a point but this does remind me of shenanigans I have had to deal with: I used to play with a guy who tried to get away with rolling the dice one at a time and if it came up low he would aim the next dice at it to try and change it.
The 1d6 plus base that I presented above is supposed to be a compromise between rolling and point buy which is why it is more like point buy. I would rather do this however, then PB because it feels random, which is more fun to me, and can still create different characters without the sameness of PB. It also makes MAD characters more viable unlike PB and traditional rolling (I do think traditional rolling is better than PB in this regard though).
Neurophage, I love your approach to things
Yep. As long as every character is on the same page, it's not a problem.
It's weird to me that rolling seems synonymous with "rolling stats in order". Very rarely have I played at tables where you couldn't assign where your stats went. When I have played games like that it was the exception and everyone was aware of the following.
1. You don't know what you're going to be able to play until after we roll.
2. We MAY not end up with a balanced party.
These two points make this extremely unappealing for anything that isn't just a one off. So usually 4d6, re-roll 1s, drop the lowest and assign the numbers however you like is the "expected norm".
That being said, I don't like point buy because I normally play full spellcasters. My issue is that as a full spell caster I MUST have a 19 in my primary casting stat just in order to fully use my class abilities. This means that while technically I can get away with having a 15 in my casting stat, it means that I have zero choice when it comes to my stat increases. I much prefer having a 17 or 18 that I can put into my casting stat before I even pick my race. The reason is that this lets me have choice. I can play a race that doesn't give a bonus to my casting stat without feeling like I'm screwing myself. When those stat increases come along it's actually meaningful because I get a choice, instead of being forced to dump the points directly into my casting stat, just so I can "advance normally" in my class.
I don't think point buy would bother me so much if I didn't have class abilities that REQUIRE me to have a 19 in a stat. Sure I can get items and enchantments in order to overcome those limitations, but now I'm having to spend resources just to "be normal", meaning that I have fewer resources I can expend on optional items/enchants.
The way that I'm used to sessions going is that everyone sort of discusses what they would like to make and then compromises are made so that we can be sure that we are covered. You know something like
"well, bob wants to play a wizard, and since no one else is rolling a healer I guess I will"
THEN
everyone starts rolling and character creation begins. Some players might already know at this point exactly which class they will be choosing, where as others might only have a vague sense of the role that they are covering and not necessarily the exact class.