Crazy Jerome
First Post
How many people do you know that are zero inches long when born?
Age starts at zero. Length starts somewhere north of that. Growth occurs in both cases.
Last edited:
Game design trap - Starting too close to zero.
- Thread starter Crazy Jerome
- Start date
How many people do you know that are zero inches long when born?
Age starts at zero. Length starts somewhere north of that. Growth occurs in both cases.
ExploderWizard
Hero
For a d20 based game, huge leaps and bounds in bonuses quickly make the math get silly.
In a d20 system, starting with a straight roll without penalty is a good starting baseline. A +4 to something at this scale is pretty huge. A bonus of greater than 6 or 8 is pushing credibility unless the goalposts are constantly moved to accomodate the bonus bloat. All this does is make the d20 roll itself more of an add on to the bonus. In this case you declare that participants must have X bonus or higher to participate leading to rigid level scaling and the consequence of individuals inhabiting the same game world to be unable to meaningfully interact with one another.
If you want a less coarse die system with the most mechanical tweaking then use percentile dice for resolution. There is a lot more room to play with when you can deal in 1% increments.
In a d20 system, starting with a straight roll without penalty is a good starting baseline. A +4 to something at this scale is pretty huge. A bonus of greater than 6 or 8 is pushing credibility unless the goalposts are constantly moved to accomodate the bonus bloat. All this does is make the d20 roll itself more of an add on to the bonus. In this case you declare that participants must have X bonus or higher to participate leading to rigid level scaling and the consequence of individuals inhabiting the same game world to be unable to meaningfully interact with one another.
If you want a less coarse die system with the most mechanical tweaking then use percentile dice for resolution. There is a lot more room to play with when you can deal in 1% increments.
Crazy Jerome
First Post
Err, no. If this had anything to do with bloat, you'd be correct, but it doesn't. You are discussing the maximum range, and I agree in a d20 system, it should be rather short. I'd prefer that the entire range be mostly concentrated in 10 numbers, with outliers growing steadily less likely for 5 numbers either side of that.
What this topic does say that you'll have more options and get better results if however big you decide to make the range, you start it at some distance from zero.
What I'd actually prefer is that the full range of bonuses be something like +6 to +25, with the "normal" range mostly +11 to +20. But most people probably wouldn't tolerate that much at start. And in any case, you do get diminising returns with this, as with anything. Getting off of +0 and +1 is a huge benefit, and it rapidly diminishes in importance from there.
Lanefan
Victoria Rules
If you want a 20-point spread as the full range of bonuses (I assume for PCs adventuring in the field) the only place where the start point of the bonus range makes a difference is in how it relates to what non-adventurers can do. Most commoner-types are likely going to be at +0 no matter what, thus if the start point of the PC bonus range is at +6 there's already a big (too big, if you ask me) gap between the lowest PC and the average person.
Other than that, does it mechanically matter if the range is +6 to +25 instead of -4 to +15, other than the optics? I suggest it doesn't, and for ease of math I'd prefer +0 to be the most common bonus encountered.
Lanefan
delericho
Legend
In a d20 based system, there's a sweet spot for modifiers around the +10 mark - that's the point where the final result on a check will be equally dependent on 'luck' and 'skill'. (Mathematically speaking, d20+10 vs DC 20 is the same as d20+50 vs DC 60... but the former feels better.) Ideally, you probably want the total modifiers used across the full campaign to be in the range -5 to +25 (at the extremes), with most of the play taking place in the +5 to +15 range.
Additionally, there's an issue with specialists and non-specialist characters, and the gap between them. If the game provides lots of options for customising characters (as in both 3e and 4e, and most likely 5e), this gap will inevitably increase with level, probably quite sharply. In 3e, it got to the point that people with a 'good' save could succeed on anything but a 1, against attacks that people with 'poor' saves needed a 20, which was suboptimal, to say the least.
That suggests to me that the correct starting point for the specialist (the Fighter with his chosen weapon) is probably about +5 at 1st level, for the non-specialist (the Cleric with a melee attack) is about +0, and the anti-specialist (the weakling Mage with non-proficient weapon) is about -5.
Then, give the characters a modest automatic increase with level. In a 20-level system, +1 per 2 levels is probably about right, while in a 30 level system +1 per 3 levels is probably better.
And then give characters plenty of options to advance those areas they care about.
That would mean that at the outset, the characters who are 'good' at something are already pretty decent, while characters who are 'bad' at something are truly lousy (as bad as the system really supports). Then, after just a few levels, characters advance into the 'sweet spot' for modifiers in the areas they specialise in.
And, ideally, as they get to the top end of the level scale, their permanent modifier in their areas of specialisation is about +20, which remains reasonably controlled, while still allowing for a bit of breathing room to add on some situational bonuses/magical buffs/whatever to take them up to the +25 that is the top end of what the system easily supports.
Additionally, there's an issue with specialists and non-specialist characters, and the gap between them. If the game provides lots of options for customising characters (as in both 3e and 4e, and most likely 5e), this gap will inevitably increase with level, probably quite sharply. In 3e, it got to the point that people with a 'good' save could succeed on anything but a 1, against attacks that people with 'poor' saves needed a 20, which was suboptimal, to say the least.
That suggests to me that the correct starting point for the specialist (the Fighter with his chosen weapon) is probably about +5 at 1st level, for the non-specialist (the Cleric with a melee attack) is about +0, and the anti-specialist (the weakling Mage with non-proficient weapon) is about -5.
Then, give the characters a modest automatic increase with level. In a 20-level system, +1 per 2 levels is probably about right, while in a 30 level system +1 per 3 levels is probably better.
And then give characters plenty of options to advance those areas they care about.
That would mean that at the outset, the characters who are 'good' at something are already pretty decent, while characters who are 'bad' at something are truly lousy (as bad as the system really supports). Then, after just a few levels, characters advance into the 'sweet spot' for modifiers in the areas they specialise in.
And, ideally, as they get to the top end of the level scale, their permanent modifier in their areas of specialisation is about +20, which remains reasonably controlled, while still allowing for a bit of breathing room to add on some situational bonuses/magical buffs/whatever to take them up to the +25 that is the top end of what the system easily supports.
Kingreaper
Adventurer
Err, no.
D20s don't really do diminishing returns, especially if you're adjusting the DCs to suit your new minimum.
Multiplicative things (like that Speed example) do diminishing returns. Damage does diminishing returns. Skill bonuses don't get diminishing returns until they start making skill checks unfailable. And +0 to +1 doesn't make anything unfailable.
% chance to hit gets diminishing returns, the same as damage, but if you're buffing the base attack bonus and the base armour, the % chance to hit won't be changed. And honestly, if fighters end up hitting AC 100% and wizards 90% (because you wanted the diminishing returns to kick in) the system is broken.
Last edited:
"I like large numbers" isn't a good reason to start things above 0, IMO, because for every one who likes them you probably find someone who dislikes them. I know I do.
Ultimately the magnitude of the bonus you add to rolls doesn't matter for the math, so the numbers should be chosen to balance "looks good" and "easy to use". I think the absence or rarity of negative numbers is enough for the former, while large numbers are bad for the latter.
With that in mind, I would balance things so that someone who is average at something has a +0 at first level. You have to be exceptionally poor for an adventurer to get a negative. With 3e/4e ability bonuses and a standard array like 16/14/13/12/10/8 that one ability you put the 8 in is the exception and even it may not take things negative (where you add something else).
This is why 3e had 4x skill points at first level, and why PF gives you a +3 bonus on trained class skills. A +4 is big when it's the maximum you can get from ability modifiers.
Ultimately the magnitude of the bonus you add to rolls doesn't matter for the math, so the numbers should be chosen to balance "looks good" and "easy to use". I think the absence or rarity of negative numbers is enough for the former, while large numbers are bad for the latter.
With that in mind, I would balance things so that someone who is average at something has a +0 at first level. You have to be exceptionally poor for an adventurer to get a negative. With 3e/4e ability bonuses and a standard array like 16/14/13/12/10/8 that one ability you put the 8 in is the exception and even it may not take things negative (where you add something else).
This is why 3e had 4x skill points at first level, and why PF gives you a +3 bonus on trained class skills. A +4 is big when it's the maximum you can get from ability modifiers.
Pyromantic
First Post
As some have already mentioned, this is completely irrelevant in the case of d20 rolls or similar mechanics that work in an additive fashion. You could apply a flat +100 bonus to all d20 rolls and increase all DCs by 100 and it would have basically no effect on the probabilities involved in the game. All d20 rolls boil down to the same thing: you are attempting to succeed on a task, and of the 20 numbers that can show on the die, a certain number of them result in success. However many of them there are is not dependent on bonuses and increases in DC that simply cancel out. There are certain corner cases that can be influenced, such as the frequency of hits that are criticals (if you always need 20 to hit, but getting 20 to hit results in a critical, then every hit is a critical) but that's about it. Appearances and ease of working with the numbers become the more important factors.
As has also been mentioned however, things that work in a more multiplicative fashion are quite different. Here the ratios become important. For example, if a target has 20 hit points and you do 4 damage per hit, then you need to connect 5 times to reduce it to 0. Get an additional +1 to damage, and now you need 4 hits. But if you only do 1 damage per hit, the +1 to damage is huge, reducing the number of hits you need from 20 to 10.
In some areas of the mechanics then, moving away from 0 as a basis makes sense, but only to the degree to which you need room for fine adjustments.
I really hope some strong mathematicians are involved in fine-tuning the mechanics for Next. Issues like this are what created problems like higher-complexity challenges actually being easier than lower-complexity in the first version of skill challenges.
As has also been mentioned however, things that work in a more multiplicative fashion are quite different. Here the ratios become important. For example, if a target has 20 hit points and you do 4 damage per hit, then you need to connect 5 times to reduce it to 0. Get an additional +1 to damage, and now you need 4 hits. But if you only do 1 damage per hit, the +1 to damage is huge, reducing the number of hits you need from 20 to 10.
In some areas of the mechanics then, moving away from 0 as a basis makes sense, but only to the degree to which you need room for fine adjustments.
I really hope some strong mathematicians are involved in fine-tuning the mechanics for Next. Issues like this are what created problems like higher-complexity challenges actually being easier than lower-complexity in the first version of skill challenges.
AbdulAlhazred
Legend
Yeah, I don't think it takes a mathematician to understand the math of an RPG (having a math degree I would have to say there's nothing in 4e math-wise that shouldn't be readily accessible to 8th grade algebra basically).
I think there are a couple separate issues here. It is true, flat linear bonuses like to-hit really can start anywhere and the ratio between +1 and +5 won't generally matter for these cases. OTOH I think the "nobody has penalties" concept is valid. It never really feels all that good to be doing something like swinging at -N. Since we can fairly well establish a base minimum weapon competency why not have that be +0? Sure, no PC is likely to be at +0, but so what? It works fine in 4e where +3 is really about the lowest bonus you ever see. Nobody bothers to even make attacks below that if they can possibly help it (IE you never see the wizard swinging his staff with a net +2 to-hit except maybe if he happens to get an OA and might as well give it a shot). I don't think a PC can even get a worst to-hit base than -1 in 4e and that's a corner case.
The top end of the scale and rate of progression are a different issue and not really part of this topic, but it seems to me that again it matters not a bit how big or small the base starting number is. Maybe the game only ever allows a max of a +20 or so, but that's a matter of progression of the power curve.
Anyway, I agree with the OP on the basic premise when it comes to things like hit points. A 2:1 ratio is a lot more reasonable than the potential 20:1 ratio that existed in AD&D (level 1 16+ CON ranger vs level 1 PC without CON bonus). The numbers could scale back a bit from 4e's numbers and it wouldn't be terrible, though I think the 4e numbers are pretty close to the ideal based on typical weapon damage and hits to knock someone down. I think a 16 hit point PC is really about the minimum that makes sense.
I think there are a couple separate issues here. It is true, flat linear bonuses like to-hit really can start anywhere and the ratio between +1 and +5 won't generally matter for these cases. OTOH I think the "nobody has penalties" concept is valid. It never really feels all that good to be doing something like swinging at -N. Since we can fairly well establish a base minimum weapon competency why not have that be +0? Sure, no PC is likely to be at +0, but so what? It works fine in 4e where +3 is really about the lowest bonus you ever see. Nobody bothers to even make attacks below that if they can possibly help it (IE you never see the wizard swinging his staff with a net +2 to-hit except maybe if he happens to get an OA and might as well give it a shot). I don't think a PC can even get a worst to-hit base than -1 in 4e and that's a corner case.
The top end of the scale and rate of progression are a different issue and not really part of this topic, but it seems to me that again it matters not a bit how big or small the base starting number is. Maybe the game only ever allows a max of a +20 or so, but that's a matter of progression of the power curve.
Anyway, I agree with the OP on the basic premise when it comes to things like hit points. A 2:1 ratio is a lot more reasonable than the potential 20:1 ratio that existed in AD&D (level 1 16+ CON ranger vs level 1 PC without CON bonus). The numbers could scale back a bit from 4e's numbers and it wouldn't be terrible, though I think the 4e numbers are pretty close to the ideal based on typical weapon damage and hits to knock someone down. I think a 16 hit point PC is really about the minimum that makes sense.
That's math that actually matters and there are people who prefer one or the other. I think the choice of whether you roll hit points should allow 5e to cover a range somewhere between 1:2 and 1:20.
For example, if typical range is from 1d6+0 to 1d12+4, the ratio is 1:3 for full hp but expands to a maximum of 1:16 if rolled.
