How to optimize

[Niro]

Well my roleplaying brothers and sisters I have decided that it is time that I learn about how to optimize.. and what it is.
Care to learn me how to do it?

Edit: In dungeons and dragons 3.5 preferably

[ShippoWildheart]

Well my roleplaying brothers and sisters I have decided that it is time that I learn about how to optimize.. and what it is.
Care to learn me how to do it?

Edit: In dungeons and dragons 3.5 preferably

1. Google a "handbook" for your class, such as "3.5 rogue handbook"
2. ???
3. Profit

[Siosilvar]

The advice is the same no matter what you're playing:

Pick something specific to do.

Pick things that make you good at doing that. Remember to compare your options. Disintegrate vs. Finger of Death? Disintegrate allows you to destroy walls, doors, undead, damage people with low Fortitude saves somewhat effectively. Finger of Death lets you kill things with low Fortitude saves instantly.

As an example:
Something to do: Make things not be able to attack (effectively).

Option 1: Wizard. Cast invisibility, darkness, solid fog, blindness, illusions of walls, web. Can't hit what you can't see or get to.

Option 2: Chain-tripper. Take Exotic Weapon Proficiency (Spiked Chain). Take Combat Reflexes. Take Improved Trip. Take Stand Still. Take other things that give you attacks of opportunities when other people do stuff. Now everything within 10 feet of you gets hit with a trip when they move. They get hit again when they fall down. Get somebody to Enlarge you and now you've got a 20ft radius.

Not the only two options, obviously.

[mostlyharmful]

There's also this as a guideline if you want to do some independant stuff beyound the guidebooks.

[dangerprawn]

Here are your first tomes.

- The Ten Commandments of Practical Optimization.
- Some handy links for CO work.
- Lists of Stuff.
- Consolidated Lists Archive.
- X stat to Y bonus.

That should keep you busy for the next two years. Come back after that.

[Zaq]

Control matters. Figure out what you want to do, and more importantly, how it can happen on your terms. Define how you're going to do things.

Then, accept that this won't always happen. In many cases, you will not be dictating the terms of how things play out exactly to your liking (unless you're a wizard with a GM who doesn't go out of their way to challenge you). Accept this, and figure out what you're going to do when your favorite tricks don't work.

Spells, naturally, are the easiest way of changing the terms of the game, but they're not the only way. That said, it really is true that full casters are simply better than anyone else.

[Starscream]

For casters: be versatile. What makes the wizard, cleric and druid the best classes is that they can do everything, usually better than the classes specifically designed to do those things. Even if you specialize as a wizard, specialize in one of the schools (transmutation, conjuration) that contain the most variable effects, and give up the ones (evocation, necromancy) that cost you the least versatility.

For non-casters: be focused. The best martial builds are those that focus on one thing, and are so good at that thing as to be godlike. Charging builds, tripping builds, grappling builds, archery builds, etc. You still won't be as powerful as the casters, but you will be better than them at one thing, and that's more than most generalist martial characters can reliably say.

There are exceptions to this. Some of the less powerful casters can be beaten by the more powerful martial characters. And there are also some jack-of-all-trades characters that split the difference. This is just a general trend.

[sonofzeal]

Synergize, synergize, synergize. Synergize like you're the Pointy Haired Boss. Make sure all your feats dovetail into your preferred technique. Consider things beyond the stereotypes, and look at how they actually affect your character. Here's an example.


Say your friend wants to play a Wizard. He look through the PHB, and see that Elves have it as a favoured class! Yay! He puts a 14 into Con, applies the stat penalty, rolls average for hitpoints, and ends up with... 19, at level 5. He also gets no real advantage over any other Wizard unless he decides to multiclass, which he probably shouldn't be doing anyway. He's also a little generic, and it's going to take a little bit of work to make him stand out from the crowd.

You're making a Wizard too. You look through the PHB, and see that Humans are a pretty good choice but you don't have any pressing need for a feat... ooo, but Dwarves get a whole slew of bonuses, and no major penalties for a Wizard! With the same stats, you end up with over 50% more hp, and a nice bonus on saves against magic. You also get a very interesting and dynamic prompt when coming up with a backstory and personality. Dwarves aren't normally Wizards; what made yours different, how does his clan see him, how does he see other Dwarves? Answering these questions will give you a leg up on making a memorable and evocative character.

[Sinfire Titan]

Here are your first tomes.

- The Ten Commandments of Practical Optimization.
- Some handy links for CO work.
- Lists of Stuff.
- Consolidated Lists Archive.
- X stat to Y bonus.

That should keep you busy for the next two years. Come back after that.

You really shouldn't link to WotC's Optimization boards these days. Those threads are barely maintained. Link to the Brilliant Gameologists ones instead. They've been maintained fairly frequently.

[Soranar]

Most optimized builds do the following:

X stat to Y

basically reduce your MAD as much as possible and pick a race that will benefit the most from this

i.e. many dwarf types have a malus to DEX but Deepwarden (PrC) lets you use your CON instead (which happens to be a bonus for you too)

being a monk or a ranger lets you use feats that you cannot get otherwise , despite not qualifying for them via stats

spellcasting

if it's limited increase it , there are as many ways to do this as there are X to Y stat options

i . e. bloodlines for sorcerers, domains for clerics, etc

make it stronger (caster level wise or via free metamagic), this is really really hard to do and limited

most of the time any class/feat that lets you do this is considered broken or too good (especially if you play a Tier 1 class) , in some rare cases it's tolerated

find counters to the usual way to handle your type of character

in short, cheat

if you melee and people fly/run out of your range, learn how to fly/teleport

become immune to the usual ways to deal with you (high saves to counter spells, evasion to help you survive failed saves, uncanny dodge to prevent sneak attacks)

Disclaimer

many options that sound good can be replaced by a magic item, this all depends on your DM (if he allows said item, if you can find one , buy one or whatever)

you should always think about this (talk it over with him) to see if that's the case

i.e. taking a bad class and multiple feats to get improved evasion when you can just buy a ring that does the same thing

if you're too successful your DM will punish you (your Pun-Pun gets killed by the gods at creation before he has time to challenge them, roll another character)

and your fellow players will probably not like you either (you steal all the kills, make them look useless in every situation, etc)

whatever contingency you invent for yourself it's probably not flawless (or DM proof) but thats the fun of the game

[Gnaeus]

Rule 1
This is a personal rule, but I try to prepare at least 3 different things that I can do in combat that are effective. Like, I will smite the enemy with my primary weapon, or if I can't do that, I will cast a spell on my enemies, or if I can't do that, I will cast a spell on an ally to make him more effective. This rule works in D&D, or almost any other game. The more different and effective the 3 options, the better. It is much harder for an enemy to neutralize a character that can do different types of actions.

This is why casters beat melee in D&D (and most other systems). A well built caster can almost always cast harmful or debilitating spells on enemies, or good ones on allies, or affect the battlefield, and many can also heal or attack with weapons or claws.

A pure fighter, on the other hand, is unlikely to have more than 2 really good options in D&D (I hit you with my primary weapon type, or I trip/grapple you). It is easy to find situations where those don't work. The prepared fighter is likely to HAVE a third option (I pull out my bow and shoot you), but he probably can't be effective at it without reducing his ability with his main attacks. Even if he can be good at melee, ranged and crowd control, those three options are a little too close for comfort, because some strategies can neutralize all three (incorporeal enemies, things you can't see, things with AC too high to hit).

Rule 2
In almost all games, combat is measured in rounds or actions. Additional rounds or actions is almost always a winning strategy to combat. Whether you do this with cyber-enhanced reflexes, or magical haste spells, or celerity, or spells or powers that can be used without taking an action, the guy with more actions usually wins, or at least gets away.

Again, Fighters fail at this in D&D. It is much easier for casters to break the action economy.

[dangerprawn]

You really shouldn't link to WotC's Optimization boards these days. Those threads are barely maintained. Link to the Brilliant Gameologists ones instead. They've been maintained fairly frequently.

All those links work. I'd have linked BG but I haven't really been on that site. It sucks about the wizards forums though. That changeover was horrible.

[PhoenixRivers]

One rule of optimization:

Get the most return for the least investment.
This applies for feats, spell effects, class features, etc. It's one of the reasons master specialist is seen as so good. It's better than wizard alone, and gives you all the prereq's for other good classes.

In battle, there are many currencies.

There's your spells available.
There's your maneuvers.
There's your movement.
There's your actions.

Of these, the most important, is actions. Action economy is the only economy. It's why a party of 4 beats a dragon. 4 actions to the dragon's 1.

It's also how you can get an edge. Anything that gives you more effective actions. Quicken spell. Twin Spell. Schism. Time Stop. Belt of Battle. All give you time, which is the most limited battle commodity.

[Doc Roc]

If you would like, the ToS chat and the ToS are good places to learn.

[Doc Roc]

There's also this as a guideline if you want to do some independant stuff beyound the guidebooks.

I think I prefer the original, without the part where a random forumite later shouts about power curves while using terrible examples. Ur-Priest is _only_ one of the strongest PrCs in the game. You know, totally behind the power curve.

:|

[Eldariel]

I think I prefer the original, without the part where a random forumite later shouts about power curves while using terrible examples. Ur-Priest is _only_ one of the strongest PrCs in the game. You know, totally behind the power curve.

:|

That link did remind me of something about why CharOps doesn't care for FAQ though:
"Very occasionally, there's simply a board-wide agreement that the rule is wrong...as with the recent FAQ claiming that Polymorph allowed the use of templated forms. "

[mostlyharmful]

I think I prefer the original, without the part where a random forumite later shouts about power curves while using terrible examples. Ur-Priest is _only_ one of the strongest PrCs in the game. You know, totally behind the power curve.

:|

so do I but my Google-fu is lazy

[sonofzeal]

I think I prefer the original, without the part where a random forumite later shouts about power curves while using terrible examples. Ur-Priest is _only_ one of the strongest PrCs in the game. You know, totally behind the power curve.

:|

Actually, I think it does exactly what he (I'm presuming he) wanted it to illustrate. Ur-Priest is devastating in 20th level builds, nobody argues that. It can, however, cause you to be objectively weaker at other levels on the way. Same with a lot of builds, really. In actual play, you need to consider how you're going to get there as well as what it'll be like once you do. Erring too much in either direction can make for a frustrating time. It doesn't matter if you'll have triple 9's at lvl 20, if you're playing a lvl 5 game.



The only flaw in the Ur-Priest example is that one of the classic entries involves Wizard straight up heading towards it, and Wizards are so far ahead of the power curve at many levels that they can afford to detour.

[Sstoopidtallkid]

That link did remind me of something about why CharOps doesn't care for FAQ though:
"Very occasionally, there's simply a board-wide agreement that the rule is wrong...as with the recent FAQ claiming that Polymorph allowed the use of templated forms. "

...yeah, every time I consider changing my sig, I hear something like that, and decide to leave it alone. They deserve it.

[ex cathedra]

It seems that I have little choice but to echo earlier posters, which isn't necessarily a bad thing. Several talented optimizers are among them.

Generally, you simply need to have a specific purpose in mind, and you need to build towards it as elegantly as you can. Several people here have mentioned handbooks. While very helpful as resources, they don't quite teach you to optimize, as much as they hand you a plate of awesome and assure you that, no matter what you pick, the resulting sandwich will, indeed, still be awesome. That's quite nice, but it isn't what you should aim for.

I'm inclined to agree that specialization is good for those without the resources to support versatility (Most martial classes go here, to be honest), while classes that can be versatile generally should be (Wizards, Erudites, Druids, and the like have this given to them; other classes, like Sorcerers, need slightly more work.)

Also, Ur-Priest is a pretty bad example, in my opinion. Sure, at level 5 or 6 it can be slightly behind, depending on the build, but it advances so quickly that any decline in power should be imperceptible by level 8 or 9.

[Doc Roc]

The only flaw in the Ur-Priest example is that one of the classic entries involves Wizard straight up heading towards it, and Wizards are so far ahead of the power curve at many levels that they can afford to detour.

Ohhhh yes, absolutely.


No. You are completely and tremendously wrong. An ur-priest theurge can get ninths on the divine side by 13th, DMM by 7th or 8th with two separate turning pools, and misses only two caster levels on the dread necromancer side. Seems really weak to me.