Base mana regeneration is just too low

Since base mana regen is based off a percentage of total mana, having more total mana will make mana regenerate faster.

More here: http://www.reddit.com/r/pathofexile/comments/17ndjg/a_look_at_mana_mechanics/

Shameless plug: mana balance calculator. Now with updated clarity progression.

Understanding is the key to making a mana build.

I don't like BM from a defense perspective, unless you're lugging around a 5k or better health pool with 4% or better regen. That was the motivation for making this sheet. I wanted to understand how to cover my mana costs.

Have a look. Which mana stats help you more, well, it depends completely on context. Could be regen, could be reduced mana. Percent increased mana always falls behind the two, though - except when it comes to supporting auras.

Love your work Zaka :)

Is the base mana regen high enough in your opinion though? You are one of the few people to have analytically reviewed how to make it work. I'd be interested to know your opinion :)

Instead of focusing on something like whether base mana regen is too low, you should look at skills individually - and decide whether you think their base mana costs make sense.

For instance, I think frenzy's base mana cost is too high. Because if you're using it right, you'll have ridiculous APS anyway; its base manacost is about 18, and because of its attack speed bonus it'll consume mana faster than elemental hit. That's kind of a contextual trade-off, though. Some skills are powerful enough that they should be impossible to support completely off regen. Some skills should require you to take glass-cannon talents, in order to link nothing but DPS supports against. Or require you to link mana leech.

Very hard to up base regen. Very easy to say something like: "too hard to make an arc build". It really is pretty hard, by the way.

Should look at an attack with very low cost, say cleave. How much regen does it take to support cleave with 4 meaningful links? Say, melee physical damage - faster attacks - life on hit. With no mana leech on gear? With 1% mana leech on gear? 2%?

Is the passive investment fair for that? To me, the investment seems a bit high - considering it's one of the cheapest skills in the game. With no leech, even 4-link cleave takes a lot of passives to cover.

Now switch and ask the same questions for an expensive skill. Like arc. It really needs to be expensive, because a 5-linked arc with nothing but DPS supports.... whew. That'd be way too good. Prolif - Faster Casting - ICD - ICS - Pen, under conductivity? Ugh. Just imagine.

This answer was probably way too long winded.

What I do believe is that hallowed mana flasks are too small and recover mana too slowly. The thing is that, the farther you progress the more expensive your skills become. It's not like you're magically getting more flask charges back for that. Basically, at very high levels of progression mana flasks aren't that amazing.

I'll agree with Frenzy by itself should not be so mana intensive given its mechanics. I plan on switching Faster Attacks out for Added Fire Damage, but now I have to contend with a 140% cost multiplier than a 115% cost multiplier. Although, taking out Faster Attacks means that I shoot slower, slowing down the attack rate and giving mana a chance to regen/leech. We'll see. I plan on experimenting with it and see if the final DPS/performance is worth it.

If GGG plans on leaving resource management as constrained as is, at least the passivetree should compensate by providing easier access to more mana-related passives that aren't located on the borders of each class, dexterity classes that is.

your reply wasn't too long at all. Good insight. Awesome spreadsheet too :)

I worked out where I am at the moment with my 3L freezing pulse.

34% uptime.
Current investment: Clarity, EB, 16.7% of nodes invested in mana/mana regen (excluding EB).
Int > 150.

Do I think 34% uptime on a 3L with 1/6th passives in mana is fair? For an offensive 3L FP? I would suggest not.

In fact, I can only get another 60% increased mana regen without going well out of my way in the witch build I have set up. That will bring me to 120%. This would bring the uptime of the same FP to 47.13%. Still not even close to 100%, and that would be with 20% ish mana node investment excluding travel.

It'd be interesting to check the actual reason, but I have a feeling that spells have larger base mana costs, but the costs were supposed to be absorbed by int granting you mana. What I would think we should be seeing is (for mana builds) attacks can be used with high uptime for Str based characters (ie, not much additional mana from passives), BUT spells with their higher base costs should still have a high uptime with the same investment for Int builds because they get extra mana through Int nodes. Currently this is not even close to the case.

EDIT: for the overbearing use of the word "Actual"
EDIT 2: made the last paragraph clear between Str and Int Builds

"

StillSingle wrote:

total mana regen incl clarity = 14.4/sec. Cost of freezing pulse per cast = 18. casting time .66 seconds.

Clearly mana regen is wrong. I don't think the mana pool is too low, but definately the mana regen is wrong.

Ninja Edit: unlinked freezing pulse is not supported even with 20% of passives invested in mana and another 20% passives in intelligence and using clarity. Considering you need passives invested in life, damage, resists, defence AS WELL, mana regen is in a bad way.

It appears you can't permanently maintain an AoE based on regen, and that seems fine to me. But is it possible to maintain that if you also use mana flasks with a sufficient amount of DPS. While you won't infinitely spam it in a boss-fight, that seems perfectly wonderful to me.

I arrived at the same conclusion when I considered this:
Base mana regen is 1.75% per second. A mana regen node gives 20%, which is 0.35%. Lower than the lowest life regen node, and life notables are 4x as good. The only node close to this is Mana Flows, which is really good, but is only a stepping stone to blood magic. Adding a base ~2 mana/sec would really help with first impressions of the game, and is about a 15-30% increase for most non-int mid-lategame builds.

Reduced mana cost passives could be good, but you can only realistically get ~30% total before you start to waste lots of points running around, and it doesn't exist on gear. They can't buff the individual nodes because getting anywhere close to 100% would be ridiculous.

They could change these nodes to Increased Mana Efficiency instead. With +100%, your skills would cost half. But even with +400%, your skills would cost 20% of their original cost. Change the reduced cost nodes to increased efficiency, probably set in line with mana regen at 20% for regular nodes and 40% for Arcane Efficiency, and possibly add it on gear/uniques. This way, these nodes are useful for most builds, and not just a niche suboptimal build bait.

On another note, Mana Flasks would be much better if they lasted longer and didn't stop when you're full. I never really understood why that was happening in the first place.

sorry for delayed response stillsingle, i have made pure ele damage rangers in the past, and ive found that with a lmp or gmp + pierce coupled with piercing shots keystone, most of the time in mob situations (which theres so many of its not funny), even with a low physical modifier, most of the time youll be able to cover 90%+ if not 100% of the mp cost of skills. sure you take a dps hit, but when you count in the piercing factor, most of the time its more than worth it in terms of sheer mob dps. try it out and see for yourself.

probably will at some stage, haven't decided what bow user to make.

Then again, I might just stick to pure life builds from now on :P BM only!! :) My witch, even when I get to 120% regen and 1000 mana can only sustain the 3L FP for 70% of the time (using Zaka's awesome spreadsheet). That is pitiful :( And the only reason it had that much up time? I didn't stack any other auras.... had 800 mana for it to chunk through before it ran out :( in and out mana is not close to right yet.

I can't see how bringing mana costs down (unless everything was 12 mana at max level) is going to make things fair between a sopell vs attack. Witch stacking mana should be able to cast kick arse spells without going dry. Especially only a 3L. What's the point if int stacking otherwise?

FlickerFlare, I have 1 Mana pot on my belt, it's ALWAYS dry. as soon as there is a blue mob, dry. as soon as there is a white mob. dry. and it doesn't stop me from running out of mana. (other 4 flasks, are 3x HP, and 1x Quicksilver).

To reiterate (I think I lost my point in the prior paragraph), witchypoo cannot kill a rare enemy before I've run a full MP flask completely dry, and had to runaway because of out of mana issues.