From 3.5 to 5e - how to optimise?

[Heliomance]

I'm an experienced 3.5 player, now in my first 5e campaign. I've got a solid handle on 3.5 optimisation, but 5e seems like a very different beast, very much stripped down and simplified. What sort of things do I want to be looking at in terms of optimisation concepts? How much of my 3.5 knowledge is transferable? What should I be looking at if I want to git gud?

I'm playing a Wizard, with a homebrew school that makes me more Rogue-like - I get a couple of skill proficiencies, proficiency with thieves' tools, light armour and a Rogue weapon prof, I think later I might start getting Expertise. I've focused on more utility-type spells over flashy combat stuff, I'm basically being an Arcane Trickster but approaching it from the Wizard side instead of the Rogue side because I wanted more magic.

[Waazraath]

I'm an experienced 3.5 player, now in my first 5e campaign. I've got a solid handle on 3.5 optimisation, but 5e seems like a very different beast, very much stripped down and simplified. What sort of things do I want to be looking at in terms of optimisation concepts? How much of my 3.5 knowledge is transferable? What should I be looking at if I want to git gud?

I'm playing a Wizard, with a homebrew school that makes me more Rogue-like - I get a couple of skill proficiencies, proficiency with thieves' tools, light armour and a Rogue weapon prof, I think later I might start getting Expertise. I've focused on more utility-type spells over flashy combat stuff, I'm basically being an Arcane Trickster but approaching it from the Wizard side instead of the Rogue side because I wanted more magic.

If you were a good optimizer in 3.5, you won't find it hard to optimize in 5e. Much less content (far less feats, spells, classes), and the game is simplified quite a bit. With a few glaring exceptions (wish/simulacrum combo, coffeelock), there isn't anything really broken. With your concept, just pick spells that fit your concept. A lot of what was good in 3.5 is still good in 5e (walls for bfc, forcecage, stuff that blocks line of sight against ranged foes at lower levels), and stuff that was less impressive in 5e (direct damage blasting) is also good in 5e, so you can't go too wrong. I'd look into races, 5e has a lot of those, if you can find something that fits your concept, and feats (war caster and/or res(con) at some point will be nice, since a lot of utility stuff is concentration).

[AdAstra]

A lot of the concepts are there, but any given rule is pretty much guaranteed to be different on some level, so reading through everything is likely to be necessary.

Just to be clear, is this character already a 5e character, or are you also looking for advice on how to port it over to that edition?

Optimization concepts that are relevant to Wizards:
-Utility and non-damagey effects remain extremely strong, but many are limited by Concentration. Blasty spells are still a great panic button/mook clearer, especially Fireball/Lightning Bolt.
-Concentration is pretty much your primary limiter. Keep in mind what spells require it, since you'll only be able to keep one of those up at a time, and that Concentration can be broken. This can be especially bad on certain spells, like Haste and Fly.
-No more Vancian casting more or less! You prepare spells, but you can cast a spell as many times as you want if you have it prepared.
-Non-cantrip spells no longer scale with caster level. However, you can "upcast" many spells by using a higher level spell slot. Only a huge benefit on some spells.
-Ritual spells can have great utility and can be ritual-cast from your spellbook for free, even if you don't have them prepared (though you do need to have them in the book). Casting a spell as a ritual does require 10 minutes in addition to the normal casting time, however.
-Infinite-use Cantrips, and the damaging ones scale with character level, and are thus unaffected by multiclassing (if anything multiclassing helps em, since you ususally get more that way). Aside from being nice backup damage, some of the utility cantrips can be quite strong. Minor Illusion is generally rated highly. Choose wisely, as you can't replace them.
-Familiars now come from a 1st level spell, Find Familiar. They're pretty great to have.
-Counterspell is now really useful, using only your reaction, though it's now a 3rd level spell.
-Armor only impedes your spellcasting if you're not proficient with it. So if you can get medium/heavy armor and/or shield proficiency, it's usually a solid idea.

There's definitely more stuff I'm forgetting, but other people will have their own piece and more specific questions will be more specifically answerable for me.

[Keravath]

5e optimization comes as often in the form of multiclass dips as anything else. Some of these though can be both thematic and in character.

For example, some people optimize wizards by taking a level of cleric to start with. You need at least 13 wisdom (14 is probably better for the extra prepared spell).

Why? By dipping a level of cleric you pick up proficiency in medium armor and a shield (some clerics get heavy armor too though that usually means you need to boost strength a lot in order to wear it efficiently). Anyway, even with medium and a shield you start with 18 AC at level 1 (scale mail+shield), you retain full spell slot progression, you get some useful 1st level cleric spells (bless, healing word + cleric school spells). For example, one level of knowledge cleric which is a popular choice, gives all of the mentioned items plus expertise in two knowledge skills (e.g. arcana and nature) which a wizard is good at.

The cost is delaying the wizard spell progression by one level for increased survivability.

----

Your character concept of an arcane trickster wizard is also one that can be acheived via multiclassing. 3 levels of arcane trickster gives you the fundamentals and then 17 levels of wizard ... if you are an elf, bladesinger might be a popular choice. You would boost dex and intelligence with this combination. Wizard spell slot progression will be two levels behind and spell selection will be 3 (with a 3 level arcane trickster dip) but you pick up cunning action, mage hand legerdemain and other neat tricks.

One of the key design features of 5e is that there are very few if any trap options. The base classes are reasonably well balanced so as long as you start with at least 14 to 16 in your primary stat, the character will be able to contribute to the party. You can optimize things in 5e and will do somewhat better than an unoptimized character but you don't need detailed knowledge of system details to get a reasonable playable character which is one of its strengths and one reason the system is more accessible.

[Biffoniacus_Furiou]

For what you described, going Rogue 2 or 3 into Wizard gets you the same result without homebrew. Or you could make a Goblin (to emulate cunning action) with the Urchin background (for the rogue skills) and go Wizard and not need any homebrew. Or just make a Bard (Lore). Probably the most optimized thing you can do is to get a familiar and have it use the help action to give you advantage on everything you do.

5e is designed around bounded accuracy, i.e. bonuses and DCs intentionally scale slowly so even a low bonus has a reasonable chance of succeeding. There aren't really any mechanics that increase your bonuses like 3.5 had, you can get proficiency, you can get expertise, you can get advantage, you can get inspiration from a bard, and you can get guidance cast on you, it takes teamwork to do all that. There's a few items that give a bonus to one specific skill check but they're very few and far between.

[Heliomance]

Multiclassing has been banned, alas, and we're all variant humans for campaign setting reasons

[Man_Over_Game]

Multiclassing has been banned, alas, and we're all variant humans for campaign setting reasons

Welp, that cuts down on a lot. Which is kind of silly, because there's really not much you can break outside of standard Paladin/Sorcerer hybrids for tables that only have 1 fight per day (as they just convert all of their spell slots into alpha-strike burst damage).

For optimization, the only thing I'd recommend at this point is picking up Resilience (Con) for better Concentration-Checks, and then just maxing out your Intelligence, then Constitution. Not much else to do beyond that.

[heavyfuel]

Honestly, if you're playing a Wizard, you're about as optimized as you can get in 5e.

There isn't a very large list of feats or spells for you to dumpster dive like there was in 3.5.

Some non-core spells are a bit better than their core counterparts, and that's it.

[Man_Over_Game]

Honestly, if you're playing a Wizard, you're about as optimized as you can get in 5e.

There isn't a very large list of feats or spells for you to dumpster dive like there was in 3.5.

Some non-core spells are a bit better than their core counterparts, and that's it.

Yeah, most of the optimization is:


Don't have more than 5 levels into any two martial classes, other than Rogue.
Try not to multiclass out of Monk if you can help it (ramps up better than most martials)
Don't multiclass out of a caster, unless you're aiming for a distinct feature (and only do so for very few levels)
Get Polearm Master, GWM, Sharpshooter, Crossbow Expert, Sentinel, whenever relevant as a Martial. Include Elven Accuracy in this list if you can reliably get Advantage on your attacks.
Get Resilience (Constitution) and ASIs as a caster.
Try to dip 1 level into Cleric if you decided you want good armor, as this impacts your spell progression less than other classes as a caster, and gives you more features to play with as a martial. This has the advantage of allowing you to take your Cleric level past level 1.


That's it. Follow those rules, and you'll basically be the most overpowered option you could be. The exception is probably Paladin/Sorcerer combos, as they can burst a day's worth of spell slots into a couple turns of attacks while rocking massive defense, but that's a very well-known exception to most rules. As a caster, you'll have to do more optimization when it comes to spell choices than you will on actually building the character.

[J-H]

There are a few feats that help martials do more damage (PAM, GWM, Sharpshooter), but less for arcane casters. Most of your "optimization" is going to be in clever or skillful use of skills, spells, and tactics to solve problems. There's a lot less "win via character sheet."

[Specter]

The single most important thing for spellcasters now is concentration. At lower levels it doesn't bother you that much, but beyond it becomes an increasing pain. You should always be preparing plenty of spells that don't require concentration, so you don't limit yourself in battle after casting your first spell.
Also taking either War Caster or Resilient (CON) is pretty much mandatory to not lose a spell as soon as you take damage.

[Segev]

If you want to dodge the homebrew, consider going Rogue and running Arcane Trickster. You're already giving up the school benefits of wizard with your build, so that loss won't be felt. Though Arcane Tricksters are behind in spell level and have fewer spells per day, so if you do want a primary caster, that's not your best bet.

A Lore Bard will also get you all sorts of skill goodies, and is a full caster in its own right with the ability to pick and choose spells from other class lists (I think they eventually get 4 off-list spells).

If you just want rogue skills, you can pick a good Background (Urchin would be my recommendation, but Criminal is also good for this, and Charlatan has potential) for them. There's a spell in Xanathar's Guide that gives temporary Expertise on a single skill, too, which might be of interest to you. If the DM is allowing Feats, taking Magic Initiate for Cleric could get you guidance, as well, which is great for a skill-monger. Doing this, you could run a standard Wizard school. I'm personally very fond of the illusionist (https://forums.giantitp.com/showthread.php?477658-Illusionist-Tricks), and think it would make for a great wizard who wants to dabble in rogue-like behaviors.

If feats are allowed, again, I'd recommend Magic Initiate for guidance with your specialized school, even. +1d4 on any skill you have a moment to cast a spell before attempting is great, and you can also use it on allies.

Optimizing spell selection, be conscious of what does and does not require Concentration. You want at least one solid combat spell you can cast that doesn't require Concentration, so that you can use a Concentration spell and then cast this other one while maintaining it. 5e made Concentration less onerous than 3e in that you can still do other things while Concentrating on a spell, but it also made so very many more spells require it as a limit on what effects can be up simultaneously. Also, your cantrips will often be your bread-and-butter. Many of them grow as you level up, and you cast them all day. It's rare for a wizard to need a weapon in 5e; possible, if he chooses no attack cantrips, but rare.

Also, look into mold earth and shape water from Xanathar's Guide to Everything; they're both really nifty cantrips.

[heavyfuel]

Also taking either War Caster or Resilient (CON) is pretty much mandatory to not lose a spell as soon as you take damage.

I'd argue that these aren't as mandatory as a lot of people make them out to be. For most Wizards, at least.

If you're a Gish who's going head first into combat, then sure. Pick either or both ASAP.

But for a Wizard that's standing in the back, away from melee attacks, away from AoE attacks (unless they are targeted at you and you alone). These Feats aren't super mandatory. I'd sooner max Int to lessen the chance I waste a precious spell slot than pick either.

[Man_Over_Game]

I'd argue that these aren't as mandatory as a lot of people make them out to be. For most Wizards, at least.

If you're a Gish who's going head first into combat, then sure. Pick either or both ASAP.

But for a Wizard that's standing in the back, away from melee attacks, away from AoE attacks (unless they are targeted at you and you alone). These Feats aren't super mandatory. I'd sooner max Int to lessen the chance I waste a precious spell slot than pick either.

I agree. They aren't "Mandatory", they're "Optimal".

Casting sustained control spells is "Optimal". Casting sustained summoning spells is "Optimal". Those are not mandatory.

Optimizing players cast those spells because they're the most powerful, and then use higher Concentration checks to improve upon them. But I've seen a lot more Evocation wizards than Conjuration, and they certainly don't care much about what their Concentration checks are and seem to do perfectly fine with just Fireball.

[heavyfuel]

I agree. They aren't "Mandatory", they're "Optimal".

That's my point. I don't think they're optimal. Not before you max Int, anyway. I'll take harder to make ST and better attack bonus any day.

Indoors it's easy to find total cover, so just get beind total cover and move out and back into as you cast (because you can split your movement in 5e). Outdoors it may not be super easy, but you can still drop prone to give ranged attacks disadvantage. Drop prone behind (non total) cover, and you'll have very good effective AC.

However, since OP is playing a VHuman, by all means she should get either as her bonus feat. I'd take Res Con if I had an odd Con score and Warcaster if it was even.

[Specter]

I'd argue that these aren't as mandatory as a lot of people make them out to be. For most Wizards, at least.

If you're a Gish who's going head first into combat, then sure. Pick either or both ASAP.

But for a Wizard that's standing in the back, away from melee attacks, away from AoE attacks (unless they are targeted at you and you alone). These Feats aren't super mandatory. I'd sooner max Int to lessen the chance I waste a precious spell slot than pick either.

Yes, I'll rephrase: they're mandatory if you want to be casting concentration spells. And it gets more and more important as you move up in levels.

In lower levels, you don't have many great concentration spells to pick, and the DC of a save if you're hit shouldn't be higher than 10, so you can do without. But in higher levels, when taking 30 or more damage is commonplace in a turn, then you need at least one of these to have a shot at concentrating decently.

[Keravath]

Yes, I'll rephrase: they're mandatory if you want to be casting concentration spells. And it gets more and more important as you move up in levels.

In lower levels, you don't have many great concentration spells to pick, and the DC of a save if you're hit shouldn't be higher than 10, so you can do without. But in higher levels, when taking 30 or more damage is commonplace in a turn, then you need at least one of these to have a shot at concentrating decently.

I'd completely agree with this. If you are casting concentration spells and expect to maintain concentration then resilient con or warcaster really are pretty much mandatory.

Although it depends to some extent on the DM, relying on not being hit because you are standing in the back, has not been a good idea in my experience. If the wizard casts an important concentration spell, then unless the opponents have very low intelligence or wisdom or simply lack the knowledge then the caster will become a primary target of any creature that can do ranged damage. Maintaining concentration with a 14 or 16 constitution on a regular basis is a challenge even against a DC of 10.

If you are playing an evoker and never cast important concentration spells then it may not matter ... any other caster I have run into benefits from improved concentration checks.

[heavyfuel]

Yes, I'll rephrase: they're mandatory if you want to be casting concentration spells. And it gets more and more important as you move up in levels.

In lower levels, you don't have many great concentration spells to pick, and the DC of a save if you're hit shouldn't be higher than 10, so you can do without. But in higher levels, when taking 30 or more damage is commonplace in a turn, then you need at least one of these to have a shot at concentrating decently.


I'd completely agree with this. If you are casting concentration spells and expect to maintain concentration then resilient con or warcaster really are pretty much mandatory.

A +3 isn't mandatory by any measure.

Neither is +1 to attack roll and save DCs, but these happen much more often than Concentration

[Heliomance]

I'm currently level 3, I took Lucky as my first level feat, and I think my background is also homebrew - pretty similar to Street Urchin though, I have plenty of skills. I almost took Skillful as my feat, but the DM was concerned I'd have too many proficiencies and would be stepping on other players' toes - I'm currently proficient in Arcana, Deception, Investigation, Perception, Sleight of Hand, Stealth, Thieves' Tools and Disguise Kits. My only direct damage spell is Ray of Frost, I'm largely taking utility spells. My spellbook currently contains:

Friends
Mage Hand
Ray of Frost

Disguise Self
Expeditious Retreat
Feather Fall
Fog Cloud
Grease
Identify Properties
Silent Image
Sleep

Detect Thoughts
Invisibility

I was very tempted by Arcane Trickster, but I wanted more spellcasting than that - I wanted to be able to play with things like Dimension Door before hitting the teens, so I decided to approach rogue-wizard hybrid from the other direction.

[AdAstra]

I'm currently level 3, I took Lucky as my first level feat, and I think my background is also homebrew - pretty similar to Street Urchin though, I have plenty of skills. I almost took Skillful as my feat, but the DM was concerned I'd have too many proficiencies and would be stepping on other players' toes - I'm currently proficient in Arcana, Deception, Investigation, Perception, Sleight of Hand, Stealth, Thieves' Tools and Disguise Kits. My only direct damage spell is Ray of Frost, I'm largely taking utility spells. My spellbook currently contains:

Friends
Mage Hand
Ray of Frost

Disguise Self
Expeditious Retreat
Feather Fall
Fog Cloud
Grease
Identify Properties
Silent Image
Sleep

Detect Thoughts
Invisibility

I was very tempted by Arcane Trickster, but I wanted more spellcasting than that - I wanted to be able to play with things like Dimension Door before hitting the teens, so I decided to approach rogue-wizard hybrid from the other direction.

Seems like a pretty nice utility spell list! The only major concerns I might have are that Expeditious Retreat and Sleep are liable to become far less useful over time (ER takes up concentration which you'll likely have better uses for, Sleep starts to get pretty rapidly outscaled by monster HP at around the level you're at). I'd also like to stress the utility provided by Find Familiar.

[Contrast]

What sort of things do I want to be looking at in terms of optimisation concepts?

Multiclassing is often more of a side-grade than an upgrade due to the delayed feats/ASIs. In fact, its generally quite difficult to build a bad character in 5E unless you start multiclassing. Optimising in 5E is very much abouot remembering the opportunity costs - oftentimes straight class is the right answer.

That said, you mentioned no multiclassing. This actually leaves you with very few choices to make honestly - subclass and ASIs/feats are it. Generally speaking, boost your main stat and you'll be fine. A couple of the feats are good but not vital.


How much of my 3.5 knowledge is transferable?

Basically none. Honestly, its probably more harmful than it is helpful. I see more rules mistakes from people who played 3.5 than I do people who are fresh to RPGs because they assume they know how something works when they instead know how it worked in a previous edition. Do not assume you know how something works, even if it has the same name. You are better off treating it as a whole new RPG and starting from scratch.


What should I be looking at if I want to git gud?

I wouldn't worry too much. The difference between the floor and ceiling for character power has been drastically reduced compared to 3.5. Honestly just playing the game will quickly teach you what works and what doesn't and as a wizard you can't really bork your character as worst comes to the worst you just scribe some new spells into your spellbook...and wizards get plenty of spells known even if you don't find scrolls honestly.

One of the most important things to get your head around as a spellcaster is concentration. Read the rules on that, review your spell list and think about how that will work in combat keeping in mind you can only have one concentration spell up at a time. You may well want to purposefully prepare spells which are worse than other spells because they don't use concentration so you can cast them while you have another spell up.

Also one of the benefits of wizards is being able to ritual cast spells without needing to prepare them so if you are given an opportunity to buy spell scrolls ritual spells are typically going to be the ones you should pick up first.

[Arkhios]

Multiclassing has been banned, alas, and we're all variant humans for campaign setting reasons

Do note that, despite many people trying their best to disagree, Rules As Written both multiclassing and feats are optional rules. As is the availability of Variant Human.

Not using an optional rule ≠ banning a basic rule. DM is the one who plays the game with the rules behind the screen. Whether or not using any or all optional rules is DM's choice just as much as it's player's choice to play a half-orc or an elf.

I'm just saying it's somewhat unfair to show a bitter face to your DM, because of that.

[Man_Over_Game]

Do note that, despite many people trying their best to disagree, Rules As Written both multiclassing and feats are optional rules. As is the availability of Variant Human.

Not using an optional rule ≠ banning a basic rule. DM is the one who plays the game with the rules behind the screen. Whether or not using any or all optional rules is DM's choice just as much as it's player's choice to play a half-orc or an elf.

I'm just saying it's somewhat unfair to show a bitter face to your DM, because of that.

At the same time, it was a decision made without the players' interests in mind. This player's concerns are directly related, not to a theme, but to a playstyle.

He basically said "I'm not allowing you to play the way you want to", and has not offered an alternative to compensate for that loss.

Unless there's info that OP isn't giving us, I'm not sure why someone shouldn't feel bitter about it. That is, unless the OP didn't discuss with their DM over what they're looking for a in a campaign. You can only fault the DM for something he'd know.

[Eldariel]

For control spells I definitely have to point you towards Pyrotechnics and Grease. They're both notable in that they're potent control effects that require no Concentration while still keeping on giving. Higher level slots also have a lot of stuff that's useful like that. Remember that in this edition, blasty spells are pretty okay too (Fireball notably) so you can easily splice few in - as a Wizard, switching stuff in and out is easy enough.

I also strongly recommend picking up Alert at some point. It's a great feat, especially on casters and Rogue-types, and you're both. A "get out of jail for free"-card when you get ambushed, a massive boost to your Initiative and enables you to hide in Fog Cloud giving all enemies Disadvantage on attacks against you.

[heavyfuel]

He basically said "I'm not allowing you to play the way you want to", and has not offered an alternative to compensate for that loss.

The DM's allowing OP to play a homebrew subclass that basically makes her a Rogue/Wizard multiclass of sorts. How is this not an alternative to compensate for that loss?

[Man_Over_Game]

The DM's allowing OP to play a homebrew subclass that basically makes her a Rogue/Wizard multiclass of sorts. How is this not an alternative to compensate for that loss?

That's a fair point. It didn't click that they were related at the time, but I realize that they weren't expressed as such. For example, could he multiclass as normal if he didn't pick the homebrew option?

Because all you'd need to make a "Roguish Wizard", one that was good at skills but better at magic, is just a normal Wizard with a single level of Rogue (likely more powerful than the multiclass). Or get a second level if you want to Hide a lot.

[Heliomance]

That's a fair point. It didn't click that they were related at the time, but I realize that they weren't expressed as such. For example, could he multiclass as normal if he didn't pick the homebrew option?

Because all you'd need to make a "Roguish Wizard", one that was good at skills but better at magic, is just a normal Wizard with a single level of Rogue (likely more powerful than the multiclass). Or get a second level if you want to Hide a lot.

Multiclass was my first idea, the DM said it would be too strong though as the benefits from splashing one level of Rogue are worth more than a delayed level of spells and feats. He homebrewed the school as a compromise.

[Contrast]

Multiclass was my first idea, the DM said it would be too strong though as the benefits from splashing one level of Rogue are worth more than a delayed level of spells and feats.

Hmm. Not sure I agree with your DM there. There isn't a lot of synergy between rogue and wizard really and depending how fast you level being a level delayed in spell progression/ASIs can be months of real world time.

Obviously a lot depends on how the homebrew is set up but if they've let you get the features you wanted without multiclassing I'd say they've probably done you a favour power wise.

[Eldariel]

Hmm. Not sure I agree with your DM there. There isn't a lot of synergy between rogue and wizard really and depending how fast you level being a level delayed in spell progression/ASIs can be months of real world time.

Obviously a lot depends on how the homebrew is set up but if they've let you get the features you wanted without multiclassing I'd say they've probably done you a favour power wise.

Yeah, it's basically the same as in 3e. The first rule of optimisation:
- Don't lose caster levels!

That's just all there is to it. The only way to get higher level spells is to take levels in casting. If you give it up, you better be already casting 9th level spells (i.e. the best thing in the game) or not really care about the optimal decision. You can get some pretty nice stuff that helps along the way with dips in some other classes but they make you worse in your primary thing, shaping reality to your whim, which is generally more powerful than anything else in the game anyways.

But yeah, sometimes Action Surge, Divine Smite or potentially even Cunning Action can be worthwhile. It's rarely as strong as the straight caster up until level 19 though, when Wizard gets an arbitrary number of 9th level spells.