D&D 3E/3.5 Jonathan Tweet: Streamlining Third Edition

The D&D 3rd Ed project was part big-picture vision and part a collection of individual decisions about rules, terms, and characters. In terms of rules, a lot of what we did amounted to streamlining.

The D&D 3rd Ed project was part big-picture vision and part a collection of individual decisions about rules, terms, and characters. In terms of rules, a lot of what we did amounted to streamlining. We removed absolute limits in favor of consequences, removed unnecessary distinctions in favor of important ones, and eliminated extraneous rules. Many of these changes seemed drastic at the time because they eliminated rules that dated back to original D&D and its first rules supplement, Greyhawk. The D&D-playing audience, however, accepted them in stride.

pic509490.jpg

Some of the work to streamline the game had already been done in the Dungeons & Dragons line (“basic” D&D or “BECMI”), and some had been done with 2E. Basic D&D offered a unified table for ability modifiers and ditched separate damage values for human-size versus large ones. 2E ditched level limits by race, level names for classes, and the awkward term “magic-user.” Both versions of the game left out attack bonuses by weapon and Armor Class, as well as the possibility that a beginning character might randomly have a suite of game-breaking psionic powers. We maintained all these changes and took these efforts further.

One overriding goal was to remove limits wherever we could. I was fond of telling players that in the new edition you could eat rocks as your rations. The players would look at me in disbelief, and I would say, “You’ll break your teeth and starve to death, but there’s no rule against eating rocks.” Likewise, there was no rule against wizards wearing armor. It hurt spellcasting, but you could do it if you wanted to. Ryan Dancey would say the same thing more succinctly: “consequences, not restrictions.”

We got rid of class and multiclass restrictions by race. At Gen Con the year before 3E released, we showed a roomful of fans an illustration of a halforc paladin, and they cheered. We also removed ability minimums and maximums for races and minimums for classes. If you wanted to play a ranger with a low Constitution, OK, you just won’t be as tough as the typical ranger. If a wizard wants to swing a sword, OK, you’re just not as skilled with it as with a quarterstaff. Was it important to say that dwarves can’t have Dexterity scores of 18? No.

We removed differences between characters that mattered least so we could focus on distinctions that mattered most. Small characters got their foot speed increased so they could keep up better with humans-size characters. Darkvision was defined as not infrared so that it didn’t implicitly give some characters the hard-to-manage ability to see heat. Druids didn’t have to fight other druids to attain high level. Paladins could have any number of magic items. Multiclassing and dual classing became the same thing instead of two quite different systems. Earlier, D&D balanced wizards by making them weak at low level and powerful at high level, but we tried to balance the classes at both low level and high level. (We failed. Spellcasters were still too good at high level.) We put all classes on the same XP table for rising in level. The original system doubly punished wizards’ hit points by giving them a lower Hit Die per level and making them lower level at any given XP total. The system also sometimes gave clerics more hit points than fighters because a cleric would be higher level than a fighter with the same XP total.

For me it was particularly satisfying to eliminate extraneous rules. We ditched percentile Strength. A big surprise was how little complaining we heard about percentile Strength going away. The fighter with 18/100 Strength was something of a icon, but players accepted the change. Percentile Strength is a rule that you don’t see other RPGs copy, and that was a pretty good sign that it wasn’t doing much for the game.

You can say the same thing for weapons dealing more or less damage again large creatures than against human-sized targets, a rule that we dropped. Personally, I loved getting rid of weapon damage values that came with bonuses, using plain dice ranges instead. A damage range of 1d6+1 became 1d8, which is pretty much the same thing. That way, every bonus added to a damage roll was a bonus that came from something other than the base weapon type—a Strength bonus, a magical bonus, or something else special. Ranged weapons lost their rate of fire. I hated the way high-Strength characters in 2E liked throwing darts (rate of fire 3/1) so that they could get their Strength bonus on damage several times. Characters became proficient in all their classes’ weapons rather than a few, and weapon specialization went away. In 2E, specialization gave the character benefits to attack rate, attack rolls, and damage rolls—effects that multiplied together to more than double the character’s average damage.

We dropped the XP bonus that characters used to get for having high ability scores. In original D&D, the only thing that a high Strength did for your character was grant them an XP bonus if they were a fighter. Strength did not affect attacks or damage. In 3E, a high Strength score did plenty for a fighter, and the XP bonus was cut as extraneous.

We let players roll Hit Dice up to 20th level rather than making them stop at 9th or 10th. In original D&D, 9th or 10th level was a sort of maximum, with spellcasters not gaining an higher-level spells thereafter. Spells of 6th to 9th level were a later addition. The system we inherited, however, went up to 20th level, and we let Hit Dice scale up to match.

In 2E, sometimes players wanted high scores and low rolls, as with thief and ranger skills or nonweapon proficiencies. Sometimes players wanted low scores and high rolls, as with THAC0, saving throws, and Armor Class. We established a system where you wanted high scores and high rolls: attacks, saving throws, and skill checks. While we were at it, we streamlined and rationalized saving throws and offered a single initiative system rather than the several systems found in 2E.

D&D is popular in part because of its legacy, so we worried that fans would object to all these changes. Overall, however, the fans ate it up. Part of the reason that we got away with big changes is that we took pains to make the new edition really feel like D&D, but that’s a topic for another essay.
 

log in or register to remove this ad

Jonathan Tweet

Jonathan Tweet

D&D 3E, Over the Edge, Everway, Ars Magica, Omega World, Grandmother Fish

dave2008

Legend
This was a mistake. Constant HP inflation has been bad for the game.

Gaining hit points every level has serious knock on effects to the game math that 2e to 5e has never successfully handled. High level combat has always been cumbersome and slow for both players and DM. To the point that a lot of groups don't even bother with high level play.

If anything, variants like E6 have shown that the Hit point per level should have been pared back 2-3 levels from 9 or 10 if anything.

From a game design perspective it would have made things a lot easier to scale in the game. 5e tried to patch the effect of HP inflation with the 'bounded accuracy' work around. But high level play has shown to still be a slog for 5e.
I think I am coming around to this viewpoint too. It was a problem in 3e/PF1; 4e; 5e; and it looks like it might be an issue in PF2e as well. I think my next 5e campaign may be an E10 version of 5e.
 

log in or register to remove this ad

This was a mistake. Constant HP inflation has been bad for the game.
In retrospect, yes, but it's not an obvious mistake. Looking at the level chart in 2E, there's obviously something weird going on around level 10. Smoothing that out, such that HP are gained at a constant rate, fixes one problem while also simplifying the underlying mechanics. That it also introduced an even-worse problem is something that would take a long time to confirm.

Likewise, the unified ability modifier reduces complexity by a significant amount, compared to the earlier slew of charts. It wasn't immediately obvious that adding +5 to a weapon damage roll, or to your HP per level roll, would be so egregiously out of line with adding +5 to a standard d20 success roll.
 

Tony Vargas

Legend
Gaining hit points every level has serious knock on effects to the game math that 2e to 5e has never successfully handled. High level combat has always been cumbersome and slow for both players and DM. To the point that a lot of groups don't even bother with high level play.
To be fair, that was the case before hp inflation from HD every level. (And, really, a critical component of that is CON /bonus/ every level, too. The peak of hp inflation was really 3e, when not only did you get 20 HD, but your CON could be boosted by level, magic books, spells & items, and every point of CON bonus was another 20 hp at 20th level...)
D&D has always had a 'sweet spot,' at which it played well, between sudden/pointless PC death and lack of options at very low level, and wildly overpowered spells & items at high level, there was a range, in some editions narrow (E6, which you mention below, being a strong indicator that, for 3.x, that sweet spot was closer to 1-6) in others broad but with attendant issues (4e treadmill) of their own.

If anything, variants like E6 have shown that the Hit point per level should have been pared back 2-3 levels from 9 or 10 if anything.
While E6 does put breaks on hps, it also, and much more significantly, stops anyone from ever gaining notoriously broken spells, like Polymorph.

5e tried to patch the effect of HP inflation with the 'bounded accuracy' work around. But high level play has shown to still be a slog for 5e.
Kinda the opposite, BA meant that creating a sense of advancement required /even more/ emphasis on hp/damage inflation over levels. Just look at the hp of some high level monsters!
 

Doug McCrae

Legend
D&D has always had a 'sweet spot,' at which it played well, between sudden/pointless PC death and lack of options at very low level, and wildly overpowered spells & items at high level, there was a range, in some editions narrow (E6, which you mention below, being a strong indicator that, for 3.x, that sweet spot was closer to 1-6) in others broad but with attendant issues (4e treadmill) of their own.
The earliest reference to a sweet spot I know of is in White Dwarf #24 (1981) in the series An Introduction to Dungeons & Dragons by Lewis Pulsipher.

D&D is most fun for third to sixth level characters, who are strong enough to adventure without fear of immediate death, strong enough to have more combat options than flight, melee, and sleep spells, but not so strong that they can laugh at monsters.​
 

Tony Vargas

Legend
The earliest reference to a sweet spot I know of is in White Dwarf #24 (1981) in the series An Introduction to Dungeons & Dragons by Lewis Pulsipher.

D&D is most fun for third to sixth level characters, who are strong enough to adventure without fear of immediate death, strong enough to have more combat options than flight, melee, and sleep spells, but not so strong that they can laugh at monsters.​
I wonder if that was in reference to D&D or AD&D? (Not that it makes a huge difference.)

(BTW, at some point, early on, I internalized level bands of 1-3, 4-7, 8-12, & 13+ as being significant (I labeled them Apprentice, Journeyman, Master, & Legendary). I do not remember, now, were I got those. The first couple may have been level ranges on modules, maybe it was just mystical numbers like 3, 7, & 13. Anyway, because I'd done that, for years, I'd say the 1e 'sweet spot' was "4-7" but, really, on reflection, it was more like 3-8, because, by 3rd, everyone was in pretty decent shape (like the quote above says), and there's no big difference between 7th & 8th level, it's not a breakpoint for much of anything, yer still casting no higher than 4th level spells, not a Lord or anything yet, etc.)
 

Jaeger

That someone better
While E6 does put breaks on hps, it also, and much more significantly, stops anyone from ever gaining notoriously broken spells, like Polymorph.

This.

It would actually take a few other steps to really fix the "high level" scaling issues..

The question is which future edition designer is brave enough to push for those changes?

Kinda the opposite, BA meant that creating a sense of advancement required /even more/ emphasis on hp/damage inflation over levels. Just look at the hp of some high level monsters!

Exactly, a solution intended to make "low level" monsters a continual threat, just compounded the problem of HP inflation.


In retrospect, yes, but it's not an obvious mistake. Looking at the level chart in 2E, there's obviously something weird going on around level 10. ... It wasn't immediately obvious that adding +5 to a weapon damage roll, or to your HP per level roll, would be so egregiously out of line with adding +5 to a standard d20 success roll.

Not Obvious? I will quibble on this...

Maybe too insular an issue for those who were not really familiar with other game systems. Which in fairness to most D&D players, D&D basically is the hobby.

But after one read through of any edition of D&D, the ramifications of expanding HP are obvious at a glance. Granted to actually parse through and find exactly when the break points occur takes a bit, but spotting the source of the problem is rather straightforward. It's been present in every edition, and caused issues at high levels in every edition.

Interesting that the 3.0 to 5e designers seemed to have never caught on to it. Given that they had the examples from previous editions actual play.

I find, more and more I am forming the opinion that they knew the problem perfectly well...

And concluded that it just didn't matter.

The idea of "high level" zero to superhero play is more important as part of the identity of D&D to the games fan-base, than having high level play be actually functional/not a slog at the table.

And so far it seems that they are right.
 

Characters became proficient in all their classes’ weapons rather than a few, and weapon specialization went away. In 2E, specialization gave the character benefits to attack rate, attack rolls, and damage rolls—effects that multiplied together to more than double the character’s average damage.

I'm a huge enthusiast of the work you did with 3e (and everywhere else, really), but I always ask myself what you're trying to achieve by nerfing the fighter while also making the wizard better.
 

Tony Vargas

Legend
It would actually take a few other steps to really fix the "high level" scaling issues..
The question is which future edition designer is brave enough to push for those changes?
After 4e?
There's "brave" and then there's foolhardy...

Not Obvious? I will quibble on this...
Maybe too insular an issue for those who were not really familiar with other game systems. Which in fairness to most D&D players, D&D basically is the hobby.
So, not obvious, to most of the folks'd be look'n, then?

It's been present in every edition, and caused issues at high levels in every edition.
See, now you're talking wizards. ;P

Seriously, though, D&D has always had issues around high level, but I can't agree that hit point inflation from HD after 'name' level was a big part of that, because, well, 1e didn't have that kind of hp inflation, and it /really/ had issues at high level, thanks to the overwhelming power gained by casters, and, 4e /didn't/ have issues with high level, in spite of having steady hit point gain of 30 levels (albeit, not HD + CON mod, just a fixed 4, 5, or 6 hp/level), though I'm sure there were folks who thought it had issues /at all levels/.

I'd point to LFQW and similar structural problems as the culprit for D&D's perennial lack of playability at high levels.

The idea of "high level" zero to superhero play is more important as part of the identity of D&D to the games fan-base, than having high level play be actually functional/not a slog at the table.
Again, I'm not sure that all correlates the way you suggest. High-level play in, say, late 3.x, wasn't necessarily a slog - it could degenerate into Rocket Tag, rather the opposite issue.
 

Alzrius

The EN World kitten
It largely did. The rules for non-human level limits were relegated to the DMG and, while characterized by being default rules, were accompanied by a substantial amount of explanation about the pros and cons of the levels limits.

I really don't think that "relegated" is the proper term for still being in one of the Core Rulebooks. Likewise, while the 2E presentation had optional rules for altering how stringently to apply them, that's a far cry from saying that they "did away" with them (or even "largely" did away with them). They weren't even that substantial, being one modest sidebar about slow advancement past the stated level limits, and another modest sidebar about exceeding level limits for demihumans with high ability scores.

Plus, most of them were raised to be at the upper edge most campaigns were likely to hit (with the notable exception of the thief).

Not really. Even leaving aside the casual use of terms like "upper edge" and "most campaigns," the degree to which the limits were raised were in most cases small, and in some cases marginal. Dwarves could be 8th level clerics in AD&D 1E, and 10th level clerics in AD&D 2E. Elves could be a 7th level fighter in AD&D 1E, and a 12th level fighter in AD&D 2E. A half-elf could be an 8th level magic-user in AD&D 1E, and a 12th level mage in AD&D 2E.

In the case of thieves, the level limit actually went down in AD&D 2E. While most demihumans could progress to an unlimited degree in the thief class in First Edition, Second Edition hit them with caps: 12th level for elves, half-elves, and dwarves, 13th level for gnomes, and 15th level for halflings.

Here are the tables from First and Second Edition, respectively:

AD&D 1E level limits.jpg
AD&D 2E level limits.jpg
 

I really don't think that "relegated" is the proper term for still being in one of the Core Rulebooks. Likewise, while the 2E presentation had optional rules for altering how stringently to apply them, that's a far cry from saying that they "did away" with them (or even "largely" did away with them). They weren't even that substantial, being one modest sidebar about slow advancement past the stated level limits, and another modest sidebar about exceeding level limits for demihumans with high ability scores.



Not really. Even leaving aside the casual use of terms like "upper edge" and "most campaigns," the degree to which the limits were raised were in most cases small, and in some cases marginal. Dwarves could be 8th level clerics in AD&D 1E, and 10th level clerics in AD&D 2E. Elves could be a 7th level fighter in AD&D 1E, and a 12th level fighter in AD&D 2E. A half-elf could be an 8th level magic-user in AD&D 1E, and a 12th level mage in AD&D 2E.

In the case of thieves, the level limit actually went down in AD&D 2E. While most demihumans could progress to an unlimited degree in the thief class in First Edition, Second Edition hit them with caps: 12th level for elves, half-elves, and dwarves, 13th level for gnomes, and 15th level for halflings.

Here are the tables from First and Second Edition, respectively:

View attachment 114908View attachment 114909
You are both right here. AD&D 2nd Edition had level limits baked in, but there was a lengthy optional chapter in the DMG which talked about ways to circumvent them. These included removing the limits (with a paragraph or two on what that would look like), making the limits uncapped but cost significantly more XP, or increase the limits in certain ways at the DM's discretion. These were all optional rules, though, so in the OP his comment sort of overlooks that this was really the first edition to flirt with the idea as opposed to actually banishing it. The BECMI edition of D&D did incorporate "ranks" for characters at level cap as a way of advancing their skills....but that was also nothing more than a weird, limiting work around.
 

Remove ads

Remove ads

Top