Tattoos: What they mean for jewels

Jewels are, for lack of a better word, boring. They're a great idea and have a decent implementation, but they feel a little dated now that we've been exposed to newer systems.

Why are jewels bad? As they're currently implemented, jewels are the most junk-generating item in the game. There are only a handful of stats possible that are actually desirable, and each of these stats has a set of affixes you need to get together for the jewel to be worth running. There is such a wide range of niche affixes that getting synergistic mods is incredibly unlikely. This makes it so there's little to no need to roll them, and finding one is more-or-less "identify and vendor" unless you luck in to one of the very few synergistic combinations. If it's not one of those combinations, then it's probably not interesting at all. That's not to say that there aren't useful mid-tier jewels, but by and large it's life, crit multi, res efficiency, and dot multi that are sought after, and the rest of the affixes are not particularly impactful.

I propose that jewels be changed to make their purpose of filling niche build issues and enabling strange mechanics more pronounced.

First, tattoos should go core. Tattoos allow a give and take on character design that allows you to trade core stats for other stats. This, by itself, is interesting enough to be a core part of PoE. Pathing on the skill tree can be altered or even designed entirely around making room for tattoos. The problem, though, is that tattoos currently compete with jewels for design space, offering similar stats.

Solution? Double down on tattoos. Make a tattoo a rollable item with a single affix that replaces a small passive of it's type with its affix. Similar to jewels, there can be four types (Red, Green, Blue, White) that correspond to each stat and a wildcard type. The larger tattoos that replace stat notables can work similarly, with their own pool of more-powerful mods and limitations.

Next, remove all affixes from jewels and add them to the tattoo affix pools, based on the type of jewel/tattoo. Crimson Jewel mods go to Crimson Tattoo mod pools, etc.

To fix jewels, you allow them to have only a prefix and a suffix. The prefix is "Allocates ___" and is rolled from the pool of notables on the tree. The suffix is a pool of notables from the Crucible weapon trees that aren't available on the skill tree by conventional pathing. These notables include mechanical alterations to skill gems, and give a permanent and extendable place for these mechanical alterations. Importantly, these are alterations and not outright buffs to the skills, lest these changes generate a similar problem to previous unique jewels that altered skills.

This, much like the tattoo changes, gives an interesting give and take to building your character. If you really want "Breath of Rime" off the tree, you can get it on a jewel, but if you want that cold mastery or the extra chance to freeze along the way, then you need to path to it.

In summary, making tattoos "core" causes a competition between tattoos and jewels, providing different solutions to the same problems in the same mechanical space. Making tattoos function similarly to the way jewels currently do, and making jewels function in a way that allows a crucible-and-anointment-like interaction with the skill tree, allows for more long-term flexibility when adding mechanics and allows more build diversity with the skill tree. It would also help cut down on the number of junk jewels by making them more generically useful to more people. Jewels could just as well be rarer as a result of these changes.

---

If you have a particularly strong opinion about this suggestion and want to see others like it, check out these:
Ancestor Totems as a full mechanic
Monster Size 5 - Walking Big Bosses.
Calamity - A hypothetical league with fighting factions that you create.

Hmm. I don't think I've ever heard this take before on jewels. I don't entirely disagree with you, but I also don't entirely agree.

Jewels are just like any other piece of gear: of COURSE they require a certain combo of stats to be usable. But to write them off completely because of this is...a bit strange. Would you say the same thing about any other piece of gear? I can get t1 life on my armor, but if all the rest of the stats are garbage...then it's garbage. Same is true of jewels.

I also don't think there are any truly useless mods on jewels. There just happen to be some that are for more niche builds, and some that are ubiquitous. But that is the same as all gear mods.

I do think they can be improved, but I don't agree that they are bad in the least. To say jewels are bad is to say that ALL gear is bad, because they are essentially the same thing.

An improvement possibility: a crafting bench for jewels maybe? No multimod or anything, but the ability to add a single prefix or suffix at will (with less power than a dropped version ofc).



Tattoos are a cool addition, but I don't think they interact, replace, or otherwise usurp the role of jewels. But in my opinion, they offer too much power. It's like cluster jewels: you either love them or you hate them because they can offer insane amounts of power, but that also makes it so that they are almost a requirement to create a functional build (thanks to GGG balancing post-cluster release). I think we might see the same thing if tattoos were to go core: the average power of builds would skyrocket and therefore GGG will come in and nerf things or buff enemies to the point where they become mandatory just to function.

Customization and options are cool and a staple of the game, sure. But TOO much customization can lead to disastrous outcomes. The passive tree is already pretty darn complicated and customizable. Throw core tattoos into the mix and it might be opening pandora's box a little too much.

Nothing can replace 4x crit multi jewels for dps in a crit build .. nothing !

"

jsuslak313 wrote:

Hmm. I don't think I've ever heard this take before on jewels. I don't entirely disagree with you, but I also don't entirely agree.

Jewels are just like any other piece of gear: of COURSE they require a certain combo of stats to be usable. But to write them off completely because of this is...a bit strange. Would you say the same thing about any other piece of gear? I can get t1 life on my armor, but if all the rest of the stats are garbage...then it's garbage. Same is true of jewels.

I also don't think there are any truly useless mods on jewels. There just happen to be some that are for more niche builds, and some that are ubiquitous. But that is the same as all gear mods.

I do think they can be improved, but I don't agree that they are bad in the least. To say jewels are bad is to say that ALL gear is bad, because they are essentially the same thing.

An improvement possibility: a crafting bench for jewels maybe? No multimod or anything, but the ability to add a single prefix or suffix at will (with less power than a dropped version ofc).



Tattoos are a cool addition, but I don't think they interact, replace, or otherwise usurp the role of jewels. But in my opinion, they offer too much power. It's like cluster jewels: you either love them or you hate them because they can offer insane amounts of power, but that also makes it so that they are almost a requirement to create a functional build (thanks to GGG balancing post-cluster release). I think we might see the same thing if tattoos were to go core: the average power of builds would skyrocket and therefore GGG will come in and nerf things or buff enemies to the point where they become mandatory just to function.

Customization and options are cool and a staple of the game, sure. But TOO much customization can lead to disastrous outcomes. The passive tree is already pretty darn complicated and customizable. Throw core tattoos into the mix and it might be opening pandora's box a little too much.

You're right that this is a bit of a weird take - I didn't do a particularly good job at expressing the problems with jewels.

Path of Exile has a critical problem in its gearing system in that certain pieces of gear require certain stats or everything else about it is junk. For example, a chest piece has to be 6-linked or the stats can't possibly be an upgrade from another 6 link. Boots basically always need move speed. You always need a certain set of resistances. This brushes up against fundamental ARPG problems with upgrading gear that I don't really want to get in to because it's somewhat of a cursed problem, but just because there are requirements doesn't mean things can't be better. For example, Path of Exile 2 is removing sockets from gear. This opens up the possibility of upgrade on chest pieces alone significantly, thus improving the overall experience.

Jewels fall in a weird place where they're incredibly powerful (see the comment by eldest-bike4 above this) or incredibly useless. In fact, they're useless far more often than they're useful, and the gap between useful and good is incredibly small. That's where the major problem is. %life and 3x crit multi is more powerful than any notable on the skill tree. But, there are 49 prefixes and and 55 suffixes on crimson jewels alone. Not counting weighting or tiers, that's a lot of luck that can't be planned around.

This is fine. It's GGG after all, right? But jewels weren't really intended to be "more power" as far as I understand it. They're meant to be an analogue to Diablo 2's charm system, which can be used to fill gaps in your build (or get more power) at the cost of inventory space. But rather than using inventory space, jewels require pathing on the tree. This is fundamental character building territory, and you can't really plan for what jewels you'll have. You can only plan for how many slots you get, at the cost of passive points. Passive points that do nothing without a jewel, which is an item that you can't plan around the stats of. This is antithetical to the idea that jewels fill gaps in your gear because you can't plan for them at all.

You posit a jewel crafting bench. This would solve the problem of not being able to plan out jewels, but it's also philosophically against GGG's principles as far as I understand. They don't want determinism, which is exactly what a crafting bench is.

Now, I want to emphasize, none of this is to say jewels are bad. They're not. I originally posit that they're boring, not bad, and I stand by that. Jewels are perfectly fine as they are in a bubble. But, the tattoo changes I propose take us out of the bubble, so let's explore that area.

You're also right that tattoos are powerful, but I definitely disagree with the mechanical competition with jewels. Tattoos offer stats like chaos resistance, %life, elemental resistances, reservation efficiency, ailment avoidance, flat damage, etc. These are largely the same stats that jewels offer. The difference is that there is an interesting mechanical tradeoff between something you get naturally (stats) and the affix you get. You get stats initially and can then trade those stats later on for others if you deem them to not be what you need. This is a direct contrast to the way jewels work where you have to actively choose to lose power (skill points) in order to take jewel slots that only offer nothing now, and only a chance of doing what you want them to do later. They also compete directly for the niche of allowing you to fill weak areas of your build.

There are more powerful mods like additional projectiles that I'm not convinced should be on regular tattoos, but could be on the notable tattoos.

I think the mod pool for tattoos should be increased to contain all of the stats that jewels have, not just some. Yes, this makes them compete even more for this niche, which is why I posit the jewel changes as well. Tattoos and jewels are perfectly capable of existing together in their current iteration without a detrimental impact to gearing, but I think the suggested changes would make jewels better, make fixing builds better, and lose nothing in the process.

With these changes, investment in a jewel socket is essentially investment in another notable on the tree. It's a wildcard, and you know you're not getting something now, but you also know the power level of what you're investing in. You might have to farm a bit to get a notable that works for you, but you at least know the value of the node is a notable, and not whatever disappointment falls out of the next monster that drops a jewel.

I can't speak to the problem of power creep. Yes, this would likely, on average, be stronger than current jewels. I think it would also drop the top-end (4x crit multi) a bit, so maybe that isn't an issue. I can say that it wouldn't be particularly UNbalanced in my opinion. Flexibility always opens more doors.

Also, FWIW, there's an interesting mechanic hidden in this where you can take a jewel socket, socket a jewel that allocates a notable stat node (like Hard Knocks or Beef), and then apply a notable tattoo to the new notable. Just a fun observation.



Thanks for the comprehensive thoughts!

"

DJFariel wrote:

Path of Exile has a critical problem in its gearing system in that certain pieces of gear require certain stats or everything else about it is junk. For example, a chest piece has to be 6-linked or the stats can't possibly be an upgrade from another 6 link. Boots basically always need move speed. You always need a certain set of resistances. This brushes up against fundamental ARPG problems with upgrading gear that I don't really want to get in to because it's somewhat of a cursed problem, but just because there are requirements doesn't mean things can't be better. For example, Path of Exile 2 is removing sockets from gear. This opens up the possibility of upgrade on chest pieces alone significantly, thus improving the overall experience.

Just responding to this before I read the whole post: I very much disagree with this take....many builds don't require a 6L chest. Many builds don't require movement speed on their boots. etc. Many builds require "different" ways to work resistances. Your judgment is clouded by your experience a bit. For every "mandatory" stat you mention, there are going to be at least 5 builds that DON'T take that mandatory stat. They might be builds you have never experienced personally. That is actually the one thing that's GREAT about the current affix system in PoE.

And they keep creating new options in builds to require less "mandatory" options. Nowadays, even life isn't completely mandatory on gear.

Okay I read your response.

You see competition, whereas I see "stacking". For example, I can get a life roll on my body armour....but I also get life rolls on every other piece of gear. That doesn't mean these "compete" with each other, but rather they bolster each other. I don't particularly see any reason that jewels and tattoos, in their current state, can't coexist in terms of functionality and interest.

You're certainly focused on jewel vs. tattoo. But all of your observations can easily be traced to every single piece of gear. They ALL share basically the same pool of mods. With very few "unique" mods only found on specific gear pieces, such as movement speed on boots. That is how the game has been built. If they were to do this differently, and have each type of gear have its own totally unique set of affixes...we would certainly not have nearly the same amount of affixes available, nor would we have the same build creativity we have now. Usable affixes would be MORE restrictive, gear across all types of characters would be MORE similar, etc.

Regarding the "deterministic" crafting bench: you are right, and yet also wrong. The current crafting bench has existed in this game for a LONG time. GGG has a problem with HIGH END deterministic crafting, but they have shown little problem with low/mid deterministic crafting. For the vast majority of the life of this game, AND in new league mechanics. The RNG is usually non-existent or minimal for the average crafting mechanic (essence, harvest, crafting bench, etc), but is ASTRONOMICAL for the higher end. A jewel bench would fall in the low/mid tier and can be created as such.

As for the jewel ideas you suggest: we had threshold jewels that did very similar things to what you are saying...and they were removed. We are starting to get more and more interesting unique jewels to play with that are rare and powerful. I think GGG's whole plan for jewels is to treat them as extra gear pieces, that may or may not be extremely strong/mid/terrible...just like 99% of all items in the game.

I just don't see JEWELS as being a necessary focus for this argument you are making. I feel like you are actually making a far bigger argument about ALL gearing in general and sort of attempting to rein it in.

Regarding the talk of what builds use what gear - Yes, there is a larger discussion to be had around gearing and progression. This discussion encompasses the concepts of progression in games with randomly-generated loot and is endemic to all ARPGs, including Path of Exile. I don't want to have that discussion because it is a problem with no solution that ultimately devolves in to reductionism and attempts to define the undefinable, esoteric, and circumstantial concept of what an upgrade is. It is a cursed problem - there is no solution that satisfies every case. The best solution is to ignore it and fix the gaps in your systems with other systems. Path of Exile does this very well. The entire idea of jewel upgrades brushes up against this conversation and, as I said, I simply don't want to have it. It's fruitless.

"

You see competition, whereas I see "stacking".

No. You're looking at this too narrowly. This isn't about getting more modifiers. This is about the intent of the systems behind the modifiers. Yes, jewels are items, but the system behind them is that I have to allocate at least one passive point to get to a jewel socket that provides me no benefit at all until I socket a jewel in it. Jewels and tattoos fundamentally affect the same system - the passive tree. While they don't directly compete for space, they thematically compete for space in that they do functionally the same thing. One just happens to do it significantly better, and that one is tattoos.

Why does it do it better? Because with tattoos you are putting a point in to a node on your skill tree and getting something for it. Then, later, you are trading what you were given for something else. This is good design. There is choice and growth associated with this mechanic. There is development of the character. There is planning that can be done.

With jewels, you don't have that. You put at least one point toward unlocking a blank, useless slot. A jewel socket does nothing for you without a jewel. Taking a jewel socket without a jewel to socket in to it is an objectively wrong choice, and that is bad design. Even if you have a jewel, there is no guarantee that the jewel you have has a power level equivalent to anything on your tree. This is also bad design.

I want to reiterate again, here, though, that jewels are not bad as they currently are in the game. They serve their purpose. However, with these tattoo changes, their purpose is diminished and could therefor be made better.

If jewels always allocate a notable that is available on the tree and some other effect, you always know that you are getting a notable's worth of power out of taking a jewel socket, even if you don't have a jewel yet. You know what the power level of that jewel socket is when you take it.

Regarding threshold jewels - yes, they previously existed and they were a terrible design too. Why? Two reasons - First, they buffed a skill in addition to mechanically changing it. This is problematic because it allowed the skill to exist in a poor state that required the jewel for the skill to be competitive for most cases. Meaning, if you wanted to play the skill, you had to also use the jewel, thus also getting the mechanical changes that came along with it, and also being required to allocate at least one jewel socket on the tree that you get no build space from. Second, they had stat placement requirements. These were poor because they further locked in a skill to only being playable if you were in certain locations on the tree. The mechanical changes for skills was not the problem with threshold jewels.

I agree with your statements on the crafting bench. The problem as it pertains to jewels doesn't apply here, though, because the band of usefulness with jewels is something like 95% useless, 4% good, 1% ridiculous. Adding the ability to craft a mod you want on to a jewel would, at best, make that something like 60% useless, 30% good, 10% ridiculous, and it's going to push way too much power in to high-end jewels. These numbers are all pulled out of my ass but that's my gut feeling on about where the spectrum lies for these.

Simply put, jewels don't need to be a part of the discussion of adding tattoos to the base game, but my suggested implementation of tattoos is a strictly-better system than jewels that parallels jewels purpose, so jewels should also be discussed and improved if able.

Jewels.

I guess I just don't agree with the entire premise then...

To call a "blank" socket, able to be filled later by something potentially worth 3+ passives, bad design is just not true for me. I think that is far more interesting than the instant gratification of a minor tattoo. Additionally, that blank socket point that is useless until you have the jewel is exactly the same as a "useless" keystone node, or a useless notable that does nothing for you until you are at a certain stage of your build. There is no difference to me: you take it when you can use it. That is the essence of the passive tree. It is even true of tattoos: slapping a 2% movement speed tattoo onto a stat that you need is also an objectively wrong choice, probably a worse choice than taking that empty jewel slot...

A hasty comparison between the two systems: tattoos are for beginners, whereas jewels and jewel slots are for more experienced players or further along builds. They require planning. But that is true of most of the passive tree. Tattoos, on the other hand, are like you said: instant swapping of one "semi-useless" thing for something that might be slightly more useful to you. But in 99% of cases, this is far less power than anything you'd get from even a moderate jewel. Tattoos are a tweak, whereas jewels are a full upgrade. Of course there are exceptions: those killer tattoos are great and comparable to a good jewel, without the need for a jewel slot. But those are rare and expensive and likely to be nerfed once someone makes some crazy 15 extra arrow build or something...

I don't see jewel slots as being boring or useless simply because they offer nothing when I take them: I see them as strategic power increases, with the player having to decide if the jewel they have is WORTH the passive point. If I don't have a jewel, I won't take the jewel slot yet....just like if I'm running CI or whatever, I wouldn't take it until I have ES. I wouldn't take an elemental node if I'm only dealing physical damage. It's all the same to me.

I totally get that you don't want to talk about gearing as a whole and trying to avoid that, but I still think even with your last response that your true problem is THAT, and not boring jewels or anything like that. Everything you describe, and the reasoning behind them is exactly the same decision making that incorporates every decision you make in-game regarding any piece of gear or any passive point allocation. This comes through especially when you are talking about jewels being garbage vs mediocre vs insane. That is literally everything in the game...thats true of tattoos as well.

Is it worth allocating/swapping gear? That's what it all boils down to. Is it worth allocating the jewel slot when you don't have a jewel: no, it does nothing for you. Is it worth slapping on a quiver when you are a melee attacker? No, it does nothing for you. Is it worth taking a node that gives you minion damage when you don't use minions? No, it does nothing for you.

I just don't see tattoos as doing anything "better" than jewels, and I don't really see jewels as having the problems that you describe. Or rather, I don't see those problems in isolation from anything else we encounter in the game. We will have to see what others have to say too because I think that's where we end up in this debate.