A Critical Thesis on PoE2 [MEGA DOC]

Chris Wilson said he'd read it Pog

4 more things I'd add at this point
>>Unlike normal action games, the sound design (as well as general I'm hit/taking dmg feedback) for getting hit OR almost hit is almost non existent yet is a big part of normal action games. The green HP bar when poisoned doesn't display remaining duration, and there's no feedback of a general "I'm taking DoT DMG" with like an arrow on your HP bar pointing down, or flashes on the HP bar when taking hits. A player character could also shout 'BLEED' when being bled.

>>Mechanics like essences don't have any context or story/questline to them and just appear out of nowhere and stay contextless throughout the game. They should all have context from the campaign to endgame when they start appearing up to the end by some mystery you solve bit by bit, not necessarily talking to an NPC after each Essence, but still mean something in relation to the games narrative. Will Abyss post 0.3 keep it's current fluff? Not to say these should have some sort of superficial NPC for each encounter, but an overarching mystery/plot + player responses would go a long way.

>>In general there is a lack of BUTTONS you press while using other buttons. Namely examples would be A. While channeling this skill, press another to imbue this skill so that it casts the other B. Perfectly time this skill then when perfectly timed press ANY skill button to release early AND use the other skill instantly C. While using this skill (Say a charge) using any other skill with perfect timing will boost this skills damage. The skill is just a trigger and has no effect on the skill itself

>>The skillpointtax (SPT) for being a Warrior, and taking, say, one mechanic from the Witch side (Minions)is say 25ish points. The SPT for taking 2 mechanics (ES AND Minions) is about 35ish. The SPT for wanting to take one mechanic from the top of the tree, and one from the bottom, is 25ish each. So to counter this, GGG puts all the relevant nodes at the start of your tree. Except this means that all the wheels get polarized towards archetypes increasingly so. To experiment with obvious combos like an Axe and a Staff at the same time, you'd need to travel a lot, pick up nodes which are overly conditional to certain spells AND balance it with everything else (including other travel nodes/your ascendancy). In general, I think the methodology by which hybridization of archetypes should not be taxed this hard by Skill Points AND on top of that you may need entirely different chase stats. There's gotta be other solutions to make this lucrative part of PoE2 more fun.

Really wish people would proof read, condense, and make sure they arent just using filler for the sake of filler in their text.


This doc is just full of copy paste AI, a bunch of unncessary filler, doesn't need to be this long and it lacks a lot of normal flow.

It even starts with question headers and they eventually get longer and longer, and then ultimately organization breaks down into just rambling.


Very hard to read unfortunately :/

So, as other people have mentioned, there's a decent amount of filler, but there is plenty of very interesting stuff and good suggestions in here, stuff I found particularly interesting:
- Commentary around refinement to enemy AI (generally).
- Adding variance to enemy AI, and making this one of the ways in which magic and rare monsters add variety to combat.
- Static nature of enemy animations / melee attacks fixing them in place with little room for rotation. Very much agreed that this is a straight up problem.
- Reduce general movement sped for mobs.
- Lack of movement abilities on the part of bosses, very little in the way of re-active dashes or backsteps (or just re-active behaviours in general).
- Having layouts where the layout is a part of what makes it challenging.
- More environmental storytelling / deterministic campaign maps.
- Destructible environments by AI. I personally think we should also be able to create and destroy bits of the environment.
- Strongly agree with the commentary around not wanting repetitive combos, so much as dynamic and re-active ability sets where different skills serve different functions.
- The whole one mistake = death is subpar, I agree. Of course, this requires taking an enormous nerf bat to recovery. I do think if we went down this route, then many of your complaints around at least life flasks becomes irrelevant.
- Passive tree needs work to make things besides damage and life / blue life valuable. More emphasis on interesting utility nodes, etc. Also hybridizing nodes that provide benefits to extremely niche scenarios.
- Changes charms to run on cooldown instead.

Since you have (indirectly) encouraged me to think, and gotten the creative juices flowing I must scream:
- Since we're on the topic of DOTA2, I think bleed should be even more like Rupture, whereby a significant portion (or all) of the damage comes from distance traveled, rather than just the act of moving. This would of course apply to the player, and we would need an audio cue / significantly more obvious visual cue than the bleed icon in the top left.
- This would have implications both for the player, and enemies, particularly enemy AI. Boss is too mobile / jumping around alot - apply a heavy bleed, inhibit those behaviours. Vice versa as a means for bosses to punish overly mobile players.
- This also gives additional utility for combinations with things that move or reposition bosses, currently knockback barely works, and I think we lost the support which changed knockback direction.
- I'm not sure what interesting things to do with poison / burning. The idea you need to consume a potion to clear poison is certainly unique. I think something more interesting for poison would be to punish you for using mana (take chaos damage in some fixed ratio relative to mana spent).
- Perhaps burning encourages movement completely counter to bleed, where the duration of the ailment is based off how much you move / dodge roll.
- Damaging ailment charms should have a threshold for activation which you can roll as a prefix / suffix. For example, limit staunching charms to only activate upon a bleed which clears 300dps for example, then there's different tiers with bleed damage ranges from 30 - 1000 for example.
- Similarly, the stunning charm should trigger based off a number of stuns in a time period, which again, the player gets to pick / dial in.
- Accuracy is a super boring stat and there's very poor visual indication / feedback as to when it is or isn't working. One idea (not saying it's a good one) of not having enough accuracy would be projectile spread, where better and better weapons require more accuracy to clean up / reduce the variance with distance. Excess accuracy could provide chance to chain or heatseeking?

"

Mashgesture#2912 написал:

This doc is just full of copy paste AI, a bunch of unncessary filler, doesn't need to be this long and it lacks a lot of normal flow.

No AI was used. I didn't have time to format everything perfectly due to covering every aspect of the game + there's a small novel's worth wordcount with 0 filler. I agree it would've been much easier to read if I formatted it completely.