PoE2 0.4 Feedback: When Design Philosophy and Systems Drift Apart

Hey GGG & fellow Player,

This comes from a place with a lot of love for the franchise, check out my signature, I changed it once in 15 years playing PoE, saying what is says after community hated on them.

That said, in general I like that the philosophy is moving towards a diverse Temple that behaves like its own ecosystem, where rooms amplify each other if you commit to interconnected setups. Conceptually, this is one of the most interesting endgame systems in PoE ever, and I played over 15k hours combined of both games.

Recent Patch and General Feedback

The problem with the Temple is that the fundamental connection logic of the Temple actively works against that philosophy, and the recent changes make this mismatch more visible rather than fixing it.

At a structural level, the Temple is built around a small number of hard chokepoints, most notably Garrison and Armoury. These are not simply “strong rooms”, they are mandatory infrastructure.

Garrison (and effectively Barracks, which are just Tier III Garrisons) acts as the central gateway into:

-Commander
-Spymaster
-Synthflesh Lab → Flesh Surgeon chains
-or their different version Legion & Transcendent Barracks

At the same time, Armoury is the only real access point into the entire "fun room" clusters:

-Golem Works
-Alchemy Lab
-Corruption Chamber
-Sacrificial Chamber
-Thaumaturge (indirectly)

What makes this especially restrictive is that Garrison and Armoury are also the only meaningful connectors to each other. They form a single narrow spine that the entire Temple ecosystem is forced through.

This already limits variety, but it becomes more problematic once hidden restrictions are taken into account.

There are non-obvious, undocumented constraints that prevent chaining rooms within the same “bubble” of the Temple.

For example:

You cannot freely chain Alchemy Labs, Corruption Chambers, Sacrificial Chambers, Generators within the same cluster. You will a) be not able to continue your chain or b) immediatly loop ending up in destruction after the Temple is closed.

You cannot combine Commander on one side of a Barracks/Garrison chain and Spymaster on the other side. Once Commander is connected, Spymaster becomes impossible to add for that chain. Again forcing you into go into one direction a) Commander Route or b) Spymaster Route which will get punished with new DR.

These restrictions further collapse the number of viable layouts. Instead of allowing organic diversity, they force players back into the same central chokepoints, increasing pressure on Garrison and Armoury even more.

The result is paradoxical:

The philosophy asks for diversity, but the system enforces repetition.

With the introduction of diminishing returns starting at four rooms of the same type, this contradiction becomes painful. The Temple still structurally requires stacking Garrison- and Armoury-adjacent rooms to function, but now doing so is explicitly penalised. This does not create alternative strategies, it simply makes the required ones feel inefficient and unsatisfying.

Juatalotli’s Medallion being restricted to Tier III rooms amplifies this further. Previously, it acted as a stabiliser for long dependency chains. Now, long chains are both:

-structurally required, and
-mechanically fragile

Even in Todays Patch 2’s reduction of destabilised rooms, while positive, does not address the core issue. Larger Temples do not automatically mean more viable paths if connection topology remains unchanged.

In practice, this leads to strategies like building long outer chains with central “buffer rooms”. The centre exists purely as sacrificial padding, not because it is interesting or synergistic. Access points exist near the chain, but cannot be connected without creating loops or violating hidden restrictions.

Ironically, allowing more connections would likely reduce abuse, not increase it. If more bidirectional or alternative connections were permitted, long linear chains would naturally collapse into loops. Loops break infinite scaling, flatten power curves, and remove the incentive to endlessly extend single chains – all without requiring harsh numerical nerfs.

My Conclusion for now, and I really hope GGG reads this, the Temple is an awesome system like Delve in PoE1:

Right now, the Temple system is constrained by its graph design, not by player behaviour. And we didn't even touch on the "different strategy" angle e.g. running Architects or Atziri over and over again without having any Chained-Temple.

As long as:

→key rooms sit behind mandatory super-nodes,
→hidden restrictions block lateral chaining within bubbles,
→and diminishing returns punish structurally required repetition,
→the system will continue to feel restrictive and unintuitive.

If the goal is genuine "Biome"-driven diversity, the solution is not further nerfs, but expanded connectivity:

-additional connectors or lift hidden restrictions,
-alternative access routes,
-or controlled loop creation.

Without this, the Temple will remain a system where players are told to diversify, but are mechanically forced into the same narrow paths, now with diminishing returns applied on top.

Choice of Strategy

Another core issue is the lack of meaningful strategic choice when engaging with the Temple.

Right now, there are effectively two alternative approaches outside of chaining:

-repeatedly running the Architect, or
-repeatedly running Atziri, while deliberately avoiding long chains.

In theory, this should represent a valid fork in playstyle. In practice, neither option functions as a true “chase strategy”.

Architects currently offer:

-a special Orb drop,
-access to specific rooms

Atziri offers:

-a burst of loot,
-a small pool of exclusive uniques.

The problem is that both strategies fully wipe or heavily destabilise the Temple, and both can be re-accessed again within 2–3 runs. As a result:

-neither choice creates long-term momentum,
-neither choice meaningfully competes with chaining,
-and neither choice creates a compelling risk vs. reward decision.

The rewards themselves are also not strong enough to justify repeatedly resetting structural progress. They are nice, but not aspirational. There is no moment where a player thinks: “I want to run the Architect now because it meaningfully improves what comes next.”

A potential direction forward would be to merge Architect interaction with chain progression, rather than treating it as a reset point. For example:

-Architect kills could unlock temporary or permanent chain-enhancing effects,
-or enable special room states that only exist after an Architect interaction,
-or introduce Temple-wide modifiers that make future chaining easier or more flexible.

Alternatively, if Architects and Atziri are meant to be true endpoints, then their rewards need to scale accordingly. They should unlock something that makes rebuilding the Temple not just acceptable, but desirable.

At the moment, both paths feel optional, but neither feels rewarding enough to be intentional for both ways of reward, direct rewards and structural rewards.

UI & UX

A second major pain point is transparency.

A few weeks ago I watched a video by Halcyonish (an upcoming creator I would highly recommend) about UI & UX design in PoE, and I strongly agree with the core message. While PoE2 made huge strides compared to PoE1 – especially with clearer terminology and systems like the dotted-line explanations for key mechanics, the Temple UI stands out as a regression.

Interacting with the Temple often feels half-blind.

There is no clear way to understand (and I'm not talking about what we figured out already with the cheat-sheet for example, I talk about anyhting besides that):

-why certain connections are allowed or blocked,
-why some chains suddenly loop while others break,
-or why specific room combinations are impossible.

More importantly, there is no in-game path to learning this.
I do not know how it works with full confidence, and I also have no reliable way to find out. Saying that, I printed over 400 Divs in the last couple of days, no flex just context.

If the answer to this is “the community will figure it out together”, that creates a contradiction:

-on one hand, the system assumes players are casual and will not research externally,
-on the other hand, when the community does figure it out and optimises it, the solutions are removed.

Ion Hazzikostas (Blizzard World of Warcraft Game Director) recently talked about this exact tension in a panel discussion: "Players will always find optimal paths. Blocking those paths instead of embracing and managing them often kills fun rather than preserving balance."

Right now, the Temple feels like a system where:

-discovery is mandatory,
-transparency is missing,
-and mastery is discouraged.

That combination is frustrating, not challenging.

Path forward & Player Mindset

Finally, there is a disconnect between player segments.

Casual players are largely unaffected by most of this. They do not read patch notes in detail, they do not follow forum discussions, and they will simply play until things “feel different”, without understanding why.

Your core players, on the other hand, do notice. And many of them will eventually quit regardless - that is normal in seasonal ARPGs. Even with multiple Mirrors in their stash, dopamine flattens out over time, and a Divine drop stops feeling special.

That makes this exact moment particularly sensitive.

I would strongly welcome a more open discussion around the Temple system. The community has now spent weeks independently exploring it, which was important. We arrived at solutions that conflict with the intended philosophy - that is okay and expected.

At this point, the ball is with GGG.

Giving players hints, examples, or reference Temples would go a long way:

-show us what a “good” Temple looks like internally,
-share layout ideas you consider healthy,
-let players rebuild, adapt, and iterate from there.

This does not require ending the league early. But opening a dialogue, offering direction, and helping especially the more casual audience engage with the Temple meaningfully would stabilise retention far more than additional nerfs. The hardcore players are gone, lets cater to all others from now on, is what I want to say.

Let us play with the system together for a bit, while you observe and iterate in the background for future updates + your regular development cycle of PoE1 & 2. Speaking of that....

Development Cycle

I want to explicitly say that I highly appreciate the current cadence at which content is being delivered across Path of Exile 1 and Path of Exile 2. From the outside, it is obvious how much work is being pushed through, and that level of output is not something I take for granted.

However, from a player perspective, it is becoming increasingly clear that this cadence is starting to work against the quality bar GGG has historically set for itself.

Path of Exile 2 0.4 was, relative to at its time for each patch GGG standards, the most visibly broken release I have experienced in over 15 years of playing PoE. Not in the sense that “some interactions did not work” - that is normal. Tooltip inaccuracies are normal. Hidden edge-case bugs, crashes, or obscure interactions are also normal in a complex ARPG.

What stood out in 0.4 was that core game issues were immediately noticeable without deep testing:

-players stuck inside obstacles,
-missing or broken floor visuals,
-disappearing textures,
-skills functioning only with WASD input but not mouse input (e.g. Flame Breath),
-Sprint still not functioning correctly on wide and ultra-wide monitors,
-the game feeling noticeably less responsive, animations or weapon swaps simply not executing on button press,
-and core mechanics that feel misleading or incorrectly implemented after minimal testing.

Even recent Abyss-related discoveries are, frankly, hard to understand from a systems perspective.

I generally do not focus on negatives, which means I am probably underreporting issues compared to other players. Yet even for someone like me, it was impossible not to notice that fundamental things were simply not working as intended.

Because similar patterns were visible in PoE1’s Keeper of Flames – a league I played extensively (40/40, full multi-mirror Valdo farmer) – this increasingly points to a time and iteration problem, not a design or talent problem.

The Temple is a perfect example of this.

The Temple system is extremely strong at its core. If I did not know better, I would assume it is simply rushed content. With one or two additional iteration cycles, most of the issues discussed above would have been identified very quickly through playtesting. The system feels like it needed just a bit more time to mature.

Had the Temple shipped in a slightly more polished state, I genuinely believe this could have been a very strong season. Instead, the narrative quickly became negative:

-“Back to Abyss, because the Temple sucks.”
-Community infighting over bug abuse and optimisation.
-Players eventually accepting the system and engaging deeply with it.
-And then, abruptly, hard reversals and restrictions that feel like punishment rather than iteration.

The Temple itself is not the problem. The Temple is genuinely insane in terms of potential.

What feels bad is the sequence:
explore → adapt → optimise → get invested → hard stop.

This raises an uncomfortable question:
Is the cost of pushing content faster than it can be properly iterated really worth the long-term impact on player trust and enjoyment?

From the outside, it feels like the Temple was very close to greatness, but released one or two iteration too early. And that is frustrating, not because the system is bad, but because it is so clearly almost there.

Thanks for reading.

@Content Creator if you see this and agree, for the love of the game, make a video about it e.g. @Raxx 2nd Patch Review Video instead of just one or @DM I can also see you talking about this and finding the right words. Do Podcasts, be respectful and manage your audience. Hate is not a way forward, and I see too much of it here in the forums and on youtube.

They have the philosophy of you cannot in everything.
Why can't all room connect to each other?
Make upgrade simple as clicking on the same room card again
You can get room removal cards by killing the architect.
For vault rooms, just let people vaal normal rooms.
There, people can specialize in anything they want whether killing atziri or building full temple.

Great writeup, I largely agree.

My frustration with the temple largely comes from:

-Difficulty/restrains to connect rooms and then maintain connections; guarded by RNG loot tiles, worsened by T3 locks now.
In large temples you cant place "shit rooms" to delete, often most tiles you draw are unusable because they would loop/paths/cant be placed.
So you need to place a tile and get it to 3+have a lock, its too hard and inconsistent.

-Paths are not just useless but dangerous in large temples, you can never place a path without looping and bricking. Agency via a passive tree to disable paths e.g. would be great.

-Golem works are too hard to get to 3, only via +1 and generator, generator only on roads; roads bad = golem works shit.

I think temple would benefit from a mechanic of allowing players to have agency over connections in a way that occasionally they could force a room to connect or block a connection that would be made.
In its current state the temple is too frustrating and restrictive to build

"

LouiX#1802 a écrit :

If the goal is genuine "Biome"-driven diversity, the solution is not further nerfs, but expanded connectivity:

-additional connectors or lift hidden restrictions,
-alternative access routes,
-or controlled loop creation.

Without this, the Temple will remain a system where players are told to diversify, but are mechanically forced into the same narrow paths, now with diminishing returns applied on top.

Great writeup & largely agree on your conclusions. You are clearly an experienced player who knows on what content to engage and which content to avoid farming.

- Tencent has a financial plan & is paying bills
- There are certain project milestones, even if the temple is 80% there, it needs to get out. Push forward, leave the confusions and problems to solve for later.
- The design is based on a game called loop hero, where there is in addition a whole dictionary and UX explanation inside the game.
- They are currently plugging up holes for huge exploit.

What remains?
The game designer is following one simple rule
"fire brigade command: Do not let new people get into the fire. Limit the access for new people (=gamer, players)". This is the Tier-3 restriction.

You summed it up: hidden restrictions -> more and further obstacles

keepers of the flame was dead league with deleted endgame, league mechanic which doesnt give loot and deletes mobs from your map, and still casualgamerdads85 (avg zizaran subscriber) liked it. no wonder you like both 3.27 circus league and 0.4

sure new league development cycle accompanied by jonathan's vision (he never stepped outside of campaign in both poe1 and poe2 and still propose some balance changes) is not harmful at all. we have atrocious 30% retention league with deleted endgame in poe1 and untested 0.4 where ggg decided to nerf adequate endgame (abyss overrun) within 24 hours but left temple untouched for 2 weeks and now gutted it

they need to get rid of vision and give us adequate poe experience, people who want dark souls with no loot must go play some other games, you are like 5% of current playerbase, go away and nothing gonna change except we wont get your harmful feedback

+1

Well written.

I think that sums it up very well.
The temple in its Foundation is a fantastic Idear and addition to the game.

- More reward diversity
- more different rooms, ( I think about Heist, Delve )
- And your Idear, making the Architekt more valuable than Punishing

I also tried to came up with some idears in this forum, but my english isnt as good as yours.

- The new patch is not a good think. This will destroy the interest in the Temple itself and kills it off.
- I talk about the Lock Medallion, that now only can Lock Tier III Rooms.
A better way to stop creating 60 Room chains, is a restritction of possible max Rooms in a chain - like 25 for example.
You than can say - now after 25 rooms, the Lock dont work anymore - only on Room 25 not further.
- Also more variety of rooms, make more possible ways of interacting with the system = More Fun = More diversity = different Loottypes like Essences
For example - Ritual rooms ( max 4. ). So u can also ran Ritual in your temple - Or essence Rooms, where u can get some of these too.
You can build your own Maplike bioms / Ecosystem .

But this Lock - Tier III room - Thing , could be a big League killer.
I mean, i was in my temple, and recognize, that something was off , than recognized "Oh i cant lock the room i placed"
Than read the Patch notes. A Patch with no warning.

The econmy starts to get wild and Prices dropped everywhere. U nearly could follow the Pricedrop in realtime. Panic sells .. for example.

Fractured orb 10div - 1h later to 2div
Some Lineage supports 160d - 1 h later - 60d
chaos to Dive ratio , get halfed in 30 Minutes.

And now what?

The problem is, that u are very right... The Temple needed more time for developement. But its a wrong decision to Cut of the accesability for Casuals , to a possible dream outcome of getting 5- 10 Div per temple.
They saw it on Youtube. They play to get the same experience - only 1 week more and i can also have a perfect temple...
And now this.
They do not nerf the Top end of players, they cut of the fun of every casual, that where interacting with the Temple, in hope to get some lines going.

A very rough and Hard, ( And i think, wrong ) Decision.

So i think - a roomchain restiction would a better decision for everyone.
This would have killed long chains, but also reward the effort, that you can still get 5-10divs per Hour. And casual got an Endgame goal, to reach this kind of level.

Meanwhile some of ur idears could be implemented midleauge and get testet, before they move on to poe1 and the new leauge there.

With Hotfix 15 released, my hope for a soft Temple overhaul is somewhat gone, especially if they dont do a manifesto or manifesto lite the next couple of days.

My gut tells me, they have no time left and need to start / continue work on PoE1 League and PoE2 0.5 + general game and will leave the temple as it is, no new room types, no new connection paths, no new Architect Role in the temple, no new restricted room reward types / re-balancing current ones etc.

They chose to not die on the hill, give us what we wanted and I'm pretty sure they keep Temple as it is, let us have fun and get 20divs per run but then revisit it for 0.6 or even 1.0.

What you think?

"

LouiX#1802 a écrit :

My gut tells me, they have no time left and need to start / continue work on PoE1 League and PoE2 0.5

Yeah, there working schedule involves Poe1 and Poe2 new endgame maps on the Atlas. There are no new rooms or room adjustments simply of time constraints for professional coder like Jonathan.

There main objective is shipping stuff and move on. Tencent is paying the bills and setting the timeline for main releases. Fixing of buggs and holes will be performed at a later stage.