Rather comprehensive feedback (not kidding), because some people were asking for it

I started playing this game about 1.5 years ago. It's come a long way since then. Many things have changed for the better, and it shows how the devs mature and see things more clearly as time goes on. Some things changed for the worse, and shown alarming signs of ideological degradation. Some wasn't changed at all, whereas it should have.

So let's review the one and a half years since the start of Open Beta. Some of this post's content has been taken directly or paraphrased from some of my earlier posts.


WHAT BECAME BETTER

* Some of the overpowered builds were brought in line, so build variety on low/mid-budget ranges is a thing now. No more 70% dual spork magic finders at the start of a league. Even though there's still a huge tendency towards fire-based builds and Spec Throw, there's at least some notable variety going on.

* Physical damage builds, and melee builds in particular, have finally received the love they deserve. Elemental damage is still good, too, so now there is some actual choice.

* More vendor recipes—that's always a very welcome thing, especially in races and early on in the leagues.

* Scion! I knew it was coming, and it's good to have a character that starts from the middle of the tree. However, the latest changes to life/ES wheels made her a lot less valuable due to pretty pathetic upper exits and the easy lateral pathway through the ES wheel.

* Vaal orb! A great item sink for legacy items, especially on the bog that is Standard economy. Seriously, it's the best orb in terms of item economy, as the rate of destroying items that way is far greater than the rate of creating more powerful versions.

* Strongboxes! Although I never intended to play Ambush (softcore just doesn't have that thing that keeps me going), the Strongbox idea is amazing and should have been realized a long time ago.

* Rogue Exiles! Love them. Wish some of them put up more of a fight outside races.

* Very pleased to see some skills given justice, such as Searing Bond and Flame Totem.

* Very pleased at every step taken for maps to become self-sustainable for solo players. As mainly a solo player, I appreciate all of it and look forward to further polishing.

* Excellent and creative new uniques such as The Three Dragons and Shackles of the Wretched. I'll take those any day over recolored rare wannabes such as Atziri's Splendour or Queen of the Forest.

* Graphics and effects, music, new areas, mtx effects, all that jazz. All ranges from good to stellar.


WHAT BECAME WORSE

The fact that imbalances are only growing in this game makes me want to take the entire balancing division and buttslap them endlessly over a robotic voice reading out those oft-quoted manifesto lines by Chris. A unique item should never be as good as a well-rolled rare, he said. GGG is against power creep, he said.

* The legacy item situation and unique item balance in general.
Needless to say that it's something that not only plagues permanent leagues immensely, but also has great effect on temp league economy.

People who have had the time and determination to accrue the needed amount of currency buy out blatantly overpowered items such as Mjolner regardless of whether they actually need them, but to have a leverage in case these items will be nerfed at some point later, still during the temp league or otherwise. This artificially inflates prices for these items. I'll point this out one agant: legacy item situation is particularly biased towards the rich who have already accumulated a number of these items, so their wealth grows exponentially compared to the rest.

Many legacy items are very strong in the current meta, enough to propel some builds into immortal status. Again, the major division here is mainly time of admittance.

The solution would be to forcefully reroll all uniques, bases, and whatnot to their current value ranges, and do this upon every patch that introduces the relevant buffs/nerfs. It would make a small percent of oldtimers pissed but will only do well for all the rest both—now and down the line. Many players who do use the legacy items regularly, like myself, understand and accept it. In no way or form are older players entitled to absurdly overpowered versions of already (mostly insanely) powerful items just because they started earlier. It will stop the ridiculous price gouging on everything that's "about to become legacy". Excuses such as "we don't want to bring the servers down for a few hours while the DB updates" are ridiculously weak; this isn't Gmail or anything else people are relying on for work or something equally urgent. Vaal orbs trashing some of these items have been the only saving grace in a long while, but clearly they aren't enough.

* Content that's supposed to be a player skill check is at best a gear check, at worst time/RNG-gated.
Is there any rational reason for Atziri to be hidden behind an item with 4-5% drop chance among a group of items (Sacrifice fragments) that share a 50/50 probability with another group of items (Vaal gems) that could only be found in a specific type of areas (Corrupted ones) that has about a 50% chance of spawning in non-map areas as of right now? Okay, a method to circumvent the randomness of area spawning has been found thankfully, but that still doesn't change the fact that I have to run content I don't find challenging, or engaging, or at all enjoyable, over and over, for a measly 1/50 chance of getting an item that'll just get me the complete set. Why? The most desperate players actually create builds particularly for farming Midnights just to reduce travel time!

Uber Atziri is even more ridiculous. I beg your pardon, it's just fucking retarded, and everyone in the game who has run her at least once or twice thinks the same. Currently players have to pay 20-24 exalts to just attempt to run the map, because they have no choice but to pay for the rarity of the last required piece that also has about 4-5% drop chance. And that axe? Another 4-5% drop chance (at best).

At least maps themselves are mostly sustainable nowadays, and are about to become fully so once Cartographer's Boxes find their way into the core game in some form.

* Insane power creep. Best-in-slot uniques existing for literally every major build meta out there by this point.
Chris, Qarl, you can preach all you want, but you've let this happen big, big time. This is truly a fuckup. It was bad enough when Lioneye's Glare was the best bow for virtually any bow build and The Searing Touch the best fire caster staff that anybody would care to attain, available starting from like normal difficulty Merveil. Now we have Atziri's Acuity and Disfavour, Windripper and Voltaxic Rift, Mjolner—oh where do I begin with this?—Crown of Eyes, Piscator's Vigil and Bino's which for their respective build meta are only second to mirror-worthy rares, and so on.

Guys, guys, please wake up. There are unique items in this game that are so fucking absurd they have no place in the game. They make already powerful builds into completely broken ones. They grant keystone-level benefits without keystone-level drawbacks. They completely ignore or outright destroy several precisely calibrated mechanics at a time—leech/regeneration rate, mana cost, reflect, cast rate... Why exactly do you even have any of these things in place if a player can just buy themself out of it with a certain item?

I don't even know what to do here. These uniques just really should have never happened. Granted, some of those can still be salvaged into balanced endgame items by reducing some of the numbers, but for items such as Soul Taker and Atziri's Acuity the damage is already done and largely irreversible.

* Low life—or rather Shavronne's Wrappings and to a lesser extent Crown of Eyes—dominates over the metagame more than ever before.
Even snapshot notwithstanding, it just grants a 30% DPS multiplier without any significant downsides that CI builds wouldn't have as well.

Solution: make a Shavronne's user more vulnerable, goddamnit. In a smart player's hands they're just walking tanks all the same. There was a video of... Akira I think?—taking a Maze Vaal slam to the face with a Shavronne's, sporting about 10k ES. And put a "Spell damage modifiers apply to attack damage at 50% efficiency" on CoE.

* Auras are now so good they're actually mandatory for every build.
Discipline is like half the ES of any ES-based build, Purity of Lightning has completely replaced armor for Lightning Coil users, Grace with IR has a similar effect for the rest—and better than Determination. Hatred provides about 25%..35% of a physical build's total DPS. Anger and Wrath contribute to at least 50% of an elemental one even with perfect flat elemental rolls on a weapon.

As if that wasn't funny enough as is, aura nodes are conventiently packed close together for the most part (only one cluster is far away, and it's one of the two least powerful ones, too). In the current metagame, a total of 102% aura efficiency is achievable, and is a core of many powerful builds. This combined with +gem level corruptions, Empower, and other crap, lead to level 20 auras behaving something like level 30. I would suggest a maximum of 74% efficiency on the tree in total, of which aura nodes per se should contribute no more than 50%. And they should probably be more dispersed around the tree rather than being on the same pathway from Ranger to Templar (Duelist is complerely fucked, for one, while Shadow/Scion are perhaps the best dedicated aura classes). A dedicated aura build has enough DPS and survivability to be strong in endgame with nearly any skill. Blood Magic keystone is absolutely useless as a result. Who cares about less mana reservation when you can have both higher DPS and survivability by running 3 auras instead and using a BM gem?

Solution: proportionally decrease the numbers on aura efficiency nodes from 79% (I believe it is currently?) to 49%. Put a hefty life multiplier behind BM keystone, like what's done with CI. Fold the reservation multiplier into the keystone itself, because at that point there's no sense wasting points—people will need every single life node on the tree to make up for the survivability handicap.

* Spectral throw is better than bows and wands.
1) You have a choice of better damage than a bow/wand build could even hope to attain if you're using a two-hander.
2) You can use 10%+ crit chance daggers with very high damage and attack speed instead of wands with shitty damage and significantly lower attack speed. You'll still have more DPS than virtually any bow build.
3) Every ST projectile effectively has built-in Pierce, Multistrike, and Increased Critical Strikes gems due to its mechanics. The fact that you only pay the mana cost only once to shoot a volley of projectiles that hit multiple enemies multiple times each AND have some weird separate crit chances that no other skill does is unbelievably powerful. It literally puts all melee skills to shame with this crap!

I can't offer an easy solution, but a further decrease to damage efficiency and an increase to mana cost at later gem levels to like Ele Hit/Frenzy/Dom Blow levels are most likely in order if you aren't going to fix the underlying mechanics.

* Lacking caps where they are due.
This problem mainly arises in the form of Cast on Critical Strike being more efficient than the vast majority of "manual" crit or non-crit builds. This is because CoC bypasses the mana cost and cast time of a hundred spells these builds have the ability to cast every second. That's pretty bad balance! The fact that mana cost attained this way is fully sustainable, they can use Vaal Pact or (for the rich boys) Atziri's Acuity to completely eliminate the problems with leech rate and reflect. Considering these builds permanently freeze everything on the screen, they have the potential for the best survivability possible because nothing ever has a chance to deal damage to them in the first place.

The other facet to this problem is that many of the numbers that need to be balanced are subject to exponential growth in the hands of a right theorycrafter. We have nuke builds that out-DPS everything and godmode builds that maintain 100% resistances to every form of damage while in action. These numbers need to be hard-capped if you want any balance to exist in the game. This would merely be a return from broken mechanics to powerful builds.

* Some skills and spells have become worse over time.
Just some examples.
— Spark definitely needs some of its damage/crit chance and duration back.
— Elemental Hit isn't powerful enough to have twice the mana cost of higher-performing skills.
— Freezing Pulse is just meh now.
— Lightning Arrow sucks without a 6th link or a GG bow, a definite sign it needs to be buffed somehow.
— Bear Trap was a fun single-target skill for budget-minded ranged characters. It became useless enough upon the introduction of global cooldowns, but the damage was also decreased for some reason, as if it was outrageous at any point at all to begin with.

* Flameblast is OP. Burn builds are OP. Proliferated burns are outrageous.
They double-dip on fire damage and resistance penetration, they have the access to insane 1H and 2H uniques that double or triple the DPS compared to a conventional rare, they largely circumvent the damage range randomness if AoE is great enough, and the base damage is much greater than needed on most of the burn-inducing skills to begin with.

With regards to Flameblast in particular, a picture is worth a thousand words. Exactly why does it have 6% crit chance with outrageous damage, and Spark/Shock Nova only 4% with absolutely measly one?

Right now every build that does spike elemental damage will benefit immensely from proliferation for either damage output (mainly fire) or survivability (mainly cold). This support gem has flown under the radar for so long despite being so poweful it's mindboggling.

Solution: tone FB down, bring cold and lightning spells up to par, tone down burn damage (it's too easy to buff), tone down proliferation effect in either time or damage effectiveness.

* Melee Splash + a single-target skill is more effective than most of the pure AoE skills.
As ludicrous as it may sound, this is the case despite the need to waste a gem link and an overall reduction of damage per hit with it. And it is applicable both to AoE range and damae in most cases, for some reason. You might want to look at Exalton's Dual Strike for a quick and easy reference. Things like this (and trigger gems of course) lead to one-skill-click spam that ultimately makes the game unengaging.

* Leech mechanics and mana sustainment without EB.
They're both particularly bad right now. Any steps to return them to usable state are welcome.


WHAT WAS BAD... AND HAD REMAINED SO

* Desync and instance crashes.
We're very much past the point where some things are acceptable. In a game balanced around instant log out, being at a server's/connection's mercy just doesn't feel good at all. I personally know many players who have quit the game because the reasons of their high-level characters' deaths were beyond their control.

Although I constantly see how devs such as Mark and Chris mention how much has been done in this area, the results suggest the complete opposite: Vaal walk took years to be fixed despite having been experienced by like every player out there, Leap Slam/Cyclone/Whirling Blades/Shield Charge are still near-unusable in tight areas, summoners pretty much never see on their client what is actually happening on the server, and Brutus is still the teleport lord in flesh. Having to rely on a dedicated "/oos" alias in HC leagues is getting absolutely tiresome. Make it so the client has more useful information to run scene calculations with. It's OVERDUE.

* Gambling vs. crafting is still as relevant as ever.
Although many steps were made in the right direction, there's still too much gambling and too little crafting. I realize that some of it is necessary for long-term addiction forming and whatnot, but this is really only relevant at very high levels of play. It is further underminded by the fact that global economy tends towards self-regulation, and so lotteries appear to cut costs and are remarkably effective at it regardless of the state of economy in any given league. They are a thing equally viable and popular in a 1-week race and in the dumpster league.

I think Onmyoji said it best in his parting post:

"

OnmyojiOmn wrote:

I wanted to build self-contained, self-sufficient characters, play them to the end of the game, and retire them. Instead, I was impeded every step of the way by a contrived gambling system that permeates every aspect of the game and as far as I can tell does nothing but drive away players. I was punished for not being lucky enough to progress without grinding, I was punished for not being tenacious enough to grind, and I was punished for not being stupid enough to play characters that weren't fun so I could play the ones that were.

Solution: introduce more vendor-recipe-assisted crafting crutches for basic and intermediate crafting needs (leveling weapons, bi-resist gear, and the like), like it was in D2. This can be a combination of existing rolling mechanics with new and existing item-specific vendor recipes (such as the "Rustic Sash + weapon = physical roll on weapon" one). This will be a huge boost to self-found, 2 hr+ race, and fresh league metagames. Introduce at least some sort of safety nets for high-level crafting (6S/6L particularly) because very few people bother to do it by hand on rare/unique pieces anyway where they can't economically leverage 20% quality to its full advantage. Why risk it when you can run a lottery? Stop gating fun content behind RNG. Stop gating fun builds behind heavily RNG-dependent items.

You guys should really take a look at how the savvy outplay your system if you want to make any progress with it. And beside that, stop making the entire drop system so economy-oriented if you aren't willing to look into how it's being manipulated at every level—and I'm not talking about botting or RMT here.

* Managing difficulty and tactical involvement.
So far the game has pursued the enemy attribute design that is best demonstrated by races such as BLAMT: dumb monsters that hit like trucks. Obviously this encourages simplistic, brutish approaches to dealing with them more than anything else; it becomes a war of attrition, a war of numbers. Instead, intelligent enemies that counteract or avoid your skills and use several different skills on their own, depending on the situation, are so much better. Remember how Half-Life's marines were a shocking contrast to Doom- and Quake-styled zombies/demons/shamblers who just blindly attacked you the moment you entered their line of sight? Smart enemies > tough enemies.

The Atziri fight is a good example of a lazily orchestrated fight. It has the elements of surprise when done the first time (the healing phase and the split phase with mirror-holding copy), but upon very brief examination for which a couple YouTube videos are fully sufficient, it devolves entirely into "get >85% fire resistance and 5.5k EHP, move away from Storm Calls and the full-stack Flameblasts (tank the small ones), don't damage the mirror-holding copy". That's it. The whole fight becomes a gear/endurance check with little else to offer. There's literally nothing else to it; it's just like Vaal Oversoul with a different skill set: move around, don't get hit by the smash, win by attrition. Similarly the second fight is largely a DPS/leech rate check where you can substitute DPS/leech rate with the amount of seconds you can maintain your Immortal Call up.

* IIQ/IIR is still very powerful in long term but completely horrible in short-term.
Magic find characters have traditionally pulled in a lot more wealth than virtually any others. Nowadays every self-respecting party contains a dedicated culler who provides better drops for everyone, and combined with party bonuses this accumulates very quickly over time.

At the same time, running a few maps such as Maelstrom of Chaos or Vaults of Atziri by a non-MF character usually isn't even worth the investment, because the drops are typically horrible. People tend to have more good loot from a single Merc Dominus run. A run that you can only benefit from, unlike a map which is a consumable item that is always an investment as it's either bought or not-sold.

For some reason certain beginner IIQ items have been nerfed while things like Windripper weren't. Go figure.

* Detonate Dead. So many things wrong with it.
1) Stays at level 1 the entire time, bypassing level/stat requirements and mana cost.
2) Requires precious little gear/passive investment to wreck stuff because it scales off of content development rather than player character's development.
3) Has built in Faster Casting in its quality bonus.
4) The only major drawback—perishable cast fuel—is now completely negated by Desecrate.

Solution: make it 10% of enemy HP at level 1 to 15% HP at level 20 with 0.25% increments per level. Nerf the cast speed, or replace it with something else entirely. Remove the physical damage part, it's beyond useless.


OTHER THINGS THAT WARRANT LOOKING INTO

* Support gem balance.
Any spell that can shotgun will inevitably be balanced around GMP and will require it to do the intended damage output. The overall increase in damage at lv20 is 2.45x. This is more damage than any other support can provide by a mile.

Any non-melee AoE skill that needs more damage (read: all of them sans Discharge perhaps) will likely benefit from Conc Effect; it's typically powerful enough to substitute two DPS supports, and there are very easily attainable ways of driving the mana costs low enough and/or cast them from life.

Any elemental skill that makes hits will benefit from Penetration gems immensely. The multiplication factor is more than 2.0x for heavily resistant enemies—and it's only those that matter anyway, since curses already take care of the rest.

This makes a perfect elemental spell automatically require a Penetration support and a GMP/Conc Effect in order to deal most of its non-crit damage. In comparison with 20/20% Increased Crit Damage: a 100 base damage boosted to 169 with Conc and mutlipled by 3.0 (taken as an average crit multi without ICD) is 507, a 100 base damage multiplied by 4.5 (3.0 + ICD) is 450. The more the overall crit multiplier, the better Conc Effect's contribution to final damage is. The only situation where you would specifically prefer ICD to Conc is a situation where you either desperately as much area as possible (what for, when you can proliferate with a lot more efficiency?), or the situation where it's basically the only source of crit multiplier increase.

With regards to cast speed, Spell Echo is better than Faster Casting on all counts. The only spells that will need Faster Casting after Spell Echo is out are 1) those that don't need two casts in a row (like Discharge), 2) those that aren't used for damage (like Lightning Warp or curses), 3) those for whom cast speed is everything (Cybil's Paw/Three Dragons Incinerate and the like).

Added Cold Damage is very bad right now. It doesn't provide enough damage to freeze enemies for any significant duration, the mana multiplier sucks horribly (needs to be like 115% to be even worth considering), and DPS-wise virtually any other support is better.

Additional Accuracy needs more accuracy to be worthwhile for any build but a wander.

Cast on Melee Kill—why is this needed at all? Most melee kills are done by AoE attacks that clears the entire coverage area more or less homogeneously. This just takes up two links for no conceivable purpose other than to cast something when there's nobody around already. Having a less than 100% cast chance and a hefty mana multiplier just adds insult to the injury.

Chance to Flee has built-in drawbacks w/r/t magic/rare/unique monsters that completely invalidate its already dubious advantage. It was only good for bleeding-based builds to begin with, because using it defensively just wasn't worth the trouble.

Fork was only good for Spark, and now that Spark is nerfed, Fork is beyond useless.

Increased Burning Damage is only useful for spells that only do damage via burn, i.e. Searing Bond and Righteous Fire.

Increased Critical Strikes is like Power Charge on Crit, except 2-3 times worse for anyone but a Power Siphon wander. Needs a lot better chance increase to be competitive with other supports in a crit build. I mean when AddAcc provides half this gem's direct numeral benefits as well as a 1.15-1.3x crit chance multiplier by the virtue of chance to hit, you know you've got a bona fide underdog support right there.

Iron Grip is very meh as a gem. Any build that invests in Str comes close enough to the keystone anyway.

Mana Leech is almost worse than Reduced Mana atm.

Minion Speed should have a negative mana multiplier to be even worth considering at the moment, definitely not 1.5. Every summoner runs Haste with like 1.5x increased efficiency these days, which comes with the added bonus of, you know, increasing the summons' DPS.

Point Blank: see Iron Grip.

Ranged Attack Totem is completely and utterly dominated by Trap.

Dual- and triple-cursing is too easy now.
The keystone is basically pointless because there are now corruptions that provide extra curses, as well as the two unique items already in wide use. Also, curses are still so amazingly effective that every character will benefit from at least two.

Wow, gonna have to bookmark and come back later. Well written so far. :)

Now if we could just get the right people at GGG to give a damn about this kind of high quality feedback. +1 OP well done.

good post

Hi

1+ for post

I suggested to Chris awhile ago about the idea of making the legacy items becoming acc bound so that they can't be traded but merely used by that player on their characters within that account. GGG would of course give like a month warning or something to the date that the item would become legacy and acc bound. I thought this would be great since it promoted some frenzied trading and then after the allotted time had passed the item wouldn't be a constant thorn in trading channels anymore.

Chris response was simply that he disliked making legacy items acc bound.

Chris please re think this idea over, business isn't always about doing things you like but doing things that are efficient and economical.

POE could also really use recipes for making orbs more craftable and less RNG, players would take the time to then:a)grind and use rng normal orbs to roll for equipment,b)save and trade or c) save orbs used in conjunction with rare items at vendor to produce refinable orbs that can be further crafted to maximize its potential via creating high lvl mod of your choice, you would just need the time to make it but instead we are left with half assed trade channel promises and a rng masochistic orb drop grindfest:(

cheers

Absolutely awesome post. I'm jealous! Makes me want to make a similar "state of the world" post which I haven't done in a long time.

I really hope GGG reads this and takes it to heart.

One of the best feedback topics I've seen in a while and I usually disagree with the majority posted here.

The only thing I'll comment on for now is regarding Atziri. She is not a gearcheck at all. I'd argue she is purely a game knowledge and skill check. You can outgear her if you want but you can also just dodge all her mechanics with adequate movement speed and enough practice.

Edit: Oh and prolif deto/flameblast builds are not really overpowered enough to need nerfing. They're very much strong but not exactly in the toptier build list.

I've been posting many of these same points, but in much less detail. Thanks for the break-down.

As someone who is taking a break from POE because of the current unbalanced/unrewarding situation that has come of it, I find it discouraging GGG hasn't addressed many of the OP's points by now.

I've been reading the patch notes, and I'm seeing things like "reduced movement speed penalty of tower shields", and I question what GGG is actually doing.

Meanwhile, things like Cast on Crit are still unplayable in group. When will GGG revamp the visual effect of CoC to allow it to perform well? Have they accepted the fact that CoC is a lag machine?

CoC/Firestorm/Spectral Throw was so much fun. It's too bad GGG doesn't create an environment to support such builds.

Cheers guys! This post took me around 4.5 hours to assemble. Probably better spent that way than theorizing some build.

I shall begin with one very simple statement (because addressing some of the things in this post which are wrong would take too much time as I have to study)

Additional Accuracy is fine. Last I checked it was in the vicinity of guaranteeing the user 85% or more accuracy, up from around 50-60% without any accuracy sources. That's an upwards of +70% DPS. And the quality bonus is great.

Edit: that is not to say that the OP does not have good points. Just that some of his points have shown to be... less than correct in the history of the game. Case-in-point Bear Trap.