I already posted these ideas here, but OP is mod and closed his own thread.
Before getting into what I'm suggesting specifically, let's get into why. Why nerf life nodes at all?
When it comes to current passive builds, life is king. Is this because monsters do scary high damage? Partly, but I have it from my sources that rebalancing monster damage is too much for GGG to fix at the moment, so let's suck that up, bury it deep down inside and never talk about it again, Taladega Nights style. Seriously, if you must complain about it, find another thread.
But I digress. Life is king, and a lot of that has to do with passive skill point opportunity cost. An 8% life node is competing with things like:
* 8% physical melee damage with specific weapon type. So you can get 8% to one damage type, when no doubt you've got some elemental tossed in there, or 8% to ALL your life. You're melee, so life is important. Life wins.
* 4% attack speed with a specific weapon type. Okay, so now we're talking about improving all damage. But it's 4% vs 8%. Hard to ignore the bigger number there. Life wins again.
* 10% more armour. This might increase your EHP against some attacks by 5% without charges, maybe even by 9% with a bunch of endurance charges. But that's some attacks, not all, and the scariest are the big hits that melt armour, and you won't get a good percentage there... and that's physical damage only, not elemental or chaos. So life wins again.
The end result is a lot of passive builds, all the good ones really, stack life like nobody's business. This isn't good for build diversity, not one bit. And to fix it, we pretty much need to either
* buff pretty much every passive node that isn't a life node, or
* nerf life nodes.
The second option is a lot easier. But how to do it without nerfing life itself?
Since we can't adjust monster damage (shake and bake!), we need to adjust base life. You know, that value that %life nodes multiply. Up that, nerf life nodes, and the end result is that the life-stacking builds we're familiar with should still work, but new, less tanky, more DPS-focused builds would be more viable, and maybe, just maybe, people wouldn't need to stack life nodes as much.
So what I'm proposing specifically is:
* all % life node values halved. 8% nodes become 4% nodes; 12% becomes 6%. The one odd-valued node, Marauder's 13%, is rounded down to 6%.
* life gained per level doubled to 12, from 6
* base life gains from passives doubled (such as shadow's +20 life node, would become +40)
* life per strength doubled to 1, from 0.5
* starting life reduced to 40 from 50
Let's go over those in order. I've already covered the first point, but the others may confuse at first.
Doubling the life gain per level is the meat and potatoes of the "buff base life" part of my suggestion. For all high-level characters, the bulk of base life is from two places: levels and life affixes on gear, each providing more than 400 base life if optimized. Since gear can't really be changed at this point (no more wipes!), doubling that 420 base life a level 70 gets from levels to 840. Big jump.
The one thing about increasing that base life is that other base life increases have less effect, leading to the next two bullet points. However, it's the second that strikes as controversial. Buffing Strength? Yep. Before, having 400 Strength gave 200 more base life, which makes less impact than you might have thought under the current system (thanks to life affixes on gear you could expect 1000 base life on a level 70 marauder before his Str bonus), but would be an even smaller effect with level-based base life doubled. So yes, I'm suggesting a Strength buff. (Yes, Dex people, I know you're hurting, but don't hate on this suggestion just because it doesn't fix that problem.)
The last change -- reducing starting life -- is a result of the Str buff. All characters start with 14 to 30 Str, so that change means adding 7 to 15 base life to level 1 characters. I don't want to make Normal Hillock too easy, and the 10-point reduction is easily made up by the level gains later.
The one thing that isn't changed is life affixes on equipment. Those stay the same, which means that, by the same logic that not buffing Strength would be a net nerf, the life affixes would receive a stealth-nerf. They'd technically do the same thing, but would have less impact relative to everything else, which got buffed.
Quick low-level example: Let's say you decide to reroll marauder and have two white +30 life coral rings in your stash to help out. You level your marauder to 3 and...
* if you go for two life nodes, you'd have 170 life with my changes, 166 currently. Not a very big difference, a little under 3% more.
* if you go for two melee damage nodes, you'd have 154 life with my changes, 137 currently. That's about 13% more life, so you were penaliized far less for choosing the melee nodes.
Higher level example: marauder, level 48, 60 points spent. You could have this build with 275% additional life currently, which would be 137% under my suggestion; let's say you've had crap luck and don't have any life affixes so gear doesn't add any, and 330 Strength.
* with the crazy life stacking, you would have 503 base life currently, and 1887 life after the life-stacking passives. With my suggestion, you'd have 946 base life, and 2243 life after the life-stacking. Significantly more!
* same situation, but now you've got some life gear, total of 250 added. Currently you'd have 753 base life, increased to 2824 life. With my suggestion, 1196 base life, increased to 2835. That's almost the same... compared to the previous result, you can see life gear will matter less, what was about a 1000-point jump before just became a 600-point jump (for this build).
* same life gear, but now half as many life passives: 137% for current, 69% post-suggestion. 753 current base life increased to 1785; 1196 suggestion base life increased to 2022. Life passives still matter, but they are just a little less mandatory; in a way skipping the life passives nullifies the effect of the life gear, it's just that this matters much more currently than it would post-suggestion.
Okay, examples complete. Thoughts?