Overarching issues of GD set design

The following post is purely subjective and touches on age-old topic of “set vs custom” power struggle.


All variety of game’s set can be classified in 4 categories:

  1. The obvious progression
  2. Glorified recoloring
  3. Generic nothing burgers
  4. Actually the ones that alter the gameplay of skills they support (a mouthfull)

The last thing to note is given that sets usually support multiple masteries, their placing in these categories also varies from mastery to mastery, i.e. X set’s Arcanist side might be interesting, while Inquisitor’s is just a recoloring.

With that out of the way I will describe said categories and my issues with them.

1. The obvious choice

These are the sets that support already established skill and damage type combo with quantitatively more power. Obvious examples are all 5-piece sets, and I actually have no issue with those even if I may consider them rather uninsipiring. The real offenders are your usual 4-pieces.
Example:
Stun Jacks is a skill that screams “lightning” and has that redundant phys part that is there only to force you into finding items with conversion. There’re are several options that allow you to convert said phys part and then theres Light’s Defender. And the only thing the set does is just more damage, one way or another. The only gameplay difference compared to using custom variance is less damage (because nothing beats bonus radius given the nature of the skill). And then the set does it for all the skills it supports. Set also has good stats (which all sets do now) so its a no brainer choice for spam SJ.
Other examples of it are Trozan’s Arcanist’s part, Infernal Knight, Chillwhisper’s NB and Inquisitor’s part and etc.
My complaint is that all these set bonuses could be relegated to single items and nothing would change - it’s just more numbers, reminiscient of D3’s design of sets giving 10000% more damage to x skill, simply not as absurd.

2. Glorified recoloring
By far the most numerous category, it speaks for itself. The skill x is by defaul blue, this amazing new set makes it green. Enjoy.
To prevent obvious counter arguments - yes, mixing and matching different skills is part of game’s identity and in 99% of the cases having them all deal the same damage type is incredibly helpful. That being said, it’s only true for cooldown casters and p :face_with_symbols_over_mouth: ts, not for a) spam casters; b) melees; Whats the point of having EoR, DE or SS in 4 colours and they all play the same is unclear to me.

3. Generic nothing burger
These are usually support sets, I actually don’t have much issues with unless their total stat value just blows everything else out of the water (i.e. Deathguard for acid).
Examples include Bloodrager (bunch of % damage and flat bleed mods for bleeing skills) and Bonemonger (rr and stats). Usually the only thing that distinguishes them from 4 custom items are the skill procs, Bonemonger’s one being pretty unique as low cd aoe nuke.

4. The good ones
They’re good because they actually change the way you play the skills they support.
Example: Ludrigan changes spam skill with almost no DoT into a cd nuke focused on DoT. Harbinger blocks all -cd mods for DB and gives it way more damage making it into a long cd devastating nuke. Runebinder allows you to play both runes as their default damage type and provide a filler skill. It can be as simple as that, really.


All in all, the reason I decided to write that homage to Gnuffster is over the last couple of patches I felt the game become increasingly more set-driven as at some point they just started provide you with everything - stats, mods, conversions. And I just happen to not see a reason for 4 already good items to have completion bonuses added on top, when those bonuses just make you deal x more damage. Sets are supposed to be special, because you need to commit 4 slots to gain same unique advantage, but they mostly don’t.

7 Likes

I mentioned this in the Radaggan thread because it removed RoK from the conduit and added to the set. It’s fun to build around odd stuff like that and puzzle your way to a solution without 3rd party tools.

In regards to your post, 2 things:

If the non set variant set up for x skill still manages to clear the highest balanced content in a acceptable timeframe safely, then I, subjectively, feel Crate succeeded.

As for your 4th classification, I wish there were more gear/transmuters that did this sort of thing. Conceptually I like the Aether PB conduit and what it does.

Crate has done a lot by redoing a bunch of MI’s to support non set play but I can’t comment as I simply do not have time to check every variant in a perspective of how I build/play.

2 Likes

last couple of patches

Brotherman this has been the case since AoM.

It’s just GD’s approach to things.

Edit: I don’t really like it either, but it is what it is. I think it stems from the fact that for unique non-set items to be powerful, they originally had to have +1 All Skills, crazy conversions, or crazy RR because interesting statlines didn’t/don’t exist in this game, and skill modifiers didn’t exist at the time. So rather than making uber-powerful individual items, we instead saw Crate make decently-powerful sets, and the design approach never really escaped from that paradigm.

2 Likes

/yaaaaaawn

Bleh, I’m going to strong disagree on the recolouring of skills grump - I actually like it and wish it was more common when using global conversions. Because it adds to the feel of the game.

Anyhow, as for the set stuff - part of that is Crate making it less likely for builds to be reliant on double rare MI’s to fill resistances/health/OA. Is it more formulaic? Sure, but it hasn’t blocked the ability to build stronk and fun builds without 4+ piece sets r.e. the top 20 list.

Also you’re forgetting newer players are going to find it easier to build with sets and have less frustration gearing up a good build if those sets are good. And reducing player friction helps keep players engaged and gain more sales via word of mouth.

Besides, insane double rares now drop a lot more regularly, to the point I get one on average every 3 SR34-35 runs and have seen them drop during farming runs in the campaign as well. So you can use at least one double rare provided it’s easy to target farm.

3 Likes

Because being a cold build but the skill being red is incredibly off putting to me.

3 Likes

Are we talking about global conversion here? Because the ability to change a skill damage type let you play different builds/class combinations, and even if it doesn’t change the skill gameplay it’s still a nice way to improve builds variety.

2 Likes

At the very least, this sort of thing is increasing the number of ways that you can play a skill and giving you the chance to use different non-set items and mastery combinations to finish the build, so it’s a plus for build diversity.

This might be the sort of thing that is less important from a min-maxing optimizing point of view, but it does allow for some fun and creative ideas. Getting to play a skill again with a different devotion setup, class combination, and damage type is something that keeps a lot of people coming back to the game.

2 Likes

If it allows for new class combinations I think these are fine. Using different support skills and items can give the build a different enough identity, even if not reflected too much in the actually playstyle.

Other than that I agree with most points, but I also feel like people generally give more feedback to set builds than non set builds (because they are easier to build) and more feedback leads generally to more buffs than nerfs

If people consequently created more non set builds and gave feedback on their performance compared to set builds, we would have different balance right now imo.

Most of the time it goes kinda like this: The non set variant is not competitive anyway, so only the set variant gets posted and then even compared to other overperforming set builds, asking for buffs to the set and thus increasing the power disparity between set and non set variant even further.

Take Demonslayer for example, it’s currently overnerfed according to testers and they say it’s worse than custom, but I haven’t seen a single actual grimtools nor a SR/CR video with actual comparisons.

5 Likes

I agree that the game has been focusing a lot on sets and that the power of the custom greens seems to have shifted over to purples every now and then (like how Pyran has recently made custom Mortar Trap kinda redundant), but I can’t say overall that the set design in GD is bad. If a set locks 4 parts of your inventory away, taking away some options and flexibility, likely also necessitating other item pieces along the way to make up for whatever the set pieces are lacking (like CC res or full conversion) you should be getting some extra power along with that to make it worth your while. If you could make a 1:1 exact same build in terms of power with 4 greens instead of 4 purples, why in the world would you ever go for the purples? Unless they provide literally everything your build could ever hope for, you’re gonna be better off getting the greens and fishing for the right affixes to get even more power that way. Which is only going to get easier with the next expansion due to rerolls. The purples take away flexibility and customisation, so they make up for it with some extra bonuses. They definitely shouldn’t be the only option and I do hope some greens that compete with sets get some of their power back that they’ve lost over the years, but I wouldn’t say the game is quite yet at the point where you’re just building by the numbers by just grabbing the one logical set for your skill or mastery combo and building around that at all times. But that might just be me cause I like to build weird.

There’s still many ways a build can come together. You can just see a combo of skills or masteries that appeals/makes sense to you and then you try to see if there’s item support to make that viable. Or you can look at a set or an item that promotes a certain skill/playstyle and try to build entirely around that set/item, at any cost. Or you simply decide you really like a skill and you want to juice it to the stratosphere with every possible bonus you can slap onto it.

Sets and custom builds can slot into that in a variety of ways depending on what setup you’re working with from the get go and any of these 4 listed set options can be the right design for the situation. If there are even really 4 categories, cause category 3 could just as easily fall within category 1, I mean how is Bloodrager any different from Ultos? Logical support within the predicted damage type? Check. Extra procs to make that a little more interesting to play? Check.

You could say that these “obvious choice” sets are kinda boring design but who gives two shits if they at least enable a playstyle that logically should be a thing but wouldn’t be without proper support. Let me put it this way, if every set were to be revolutionary, completely changing how a skill plays, then where exactly would the game’s baseline even be? If you start playing the game and you see lightning Primal Strike, the logical conclusion that you’d expect to arrive at at endgame shouldn’t be “we have Acid Primal Strike that’s enabled by a caster offhand, Aether Primal Strike with a gun, and jellybean Primal Strike that only damages enemies that are not on your screen, and you need to do a handstand to enable it”. You want there to be some basic bitch support for Primal Strike so that you can play it the way it was originally designed. With a chonky two-handed weapon, and lightning. Sure Ultos is dull but at least it’s there to let Lightning PS be a thing (almost all Lightning PS custom items compete for the weapon slot with most other slots having nothing to build with), and it spices things up a little more than an MI would by bringing in a few big procs along the way. You couldn’t just slap that massive lightning storm on a single MI item. You need some inventory commitment, otherwise any build that even slightly smells of Primal Strike would use that massive lightning storm MI. So yeah, sure Ultos isn’t revolutionary. But at least it’s there for those who liked that core playstyle during levelling, and brings in some extra flavour to boot to make you feel like a thunder god. As much as you may scoff at how basic the design of the set it, what would some skills give to at least have that. Like 2H ranged Fire Fire Strike, for example, one of the most basic and logical options for Fire Strike, that has support all the way through levelling through stuff like Flame Keeper or Brimstone and then exactly zero support in lategame. That build would give anything to at least have a basic bitch set, cause the MIs for it aren’t there either. And it would probably prefer a set over MIs because of the aforementioned ability to add cool procs to spice up a potentially dull playstyle.

As others have mentioned and you yourself said, the game is all about the ability to combine things together in different ways, so I really don’t understand the reason to complain about skill conversion sets. That’s no fucking glorified recolouring, what the hell are you on about? Are you actually trying to tell me that EoR in Chaos and EoR in Fire are the exact same thing? Why? Cause both times you’re just holding a button to channel? So is FoI only ever allowed to have one build because they play the same? Should we get rid of the transmuter on AAR, cause you’re still just holding one button? Cool, that’s some lovely build variety we’re promoting. Do we really need a set to put a cooldown on Cadence with every other Cadence-related set so that the gameplay experience is sufficiently transformative for you? You need basic bitch gear support, conversions or not, else the endgame devolves into a slew of funky, illogical playstyles, and every core playstyle the game has been promoting during levelling ends up dead in the water come lategame. And if you want to argue that wacky conversion for skills is fine, it just shouldn’t eat up an entire set, I would encourage you to find custom item support for Aether Blitz if you delete the entirety of Krieg. That’s how it would go for set-supported conversion builds if you just slapped that conversion on one MI instead and threw the build into the wild. They’d essentially just turn into Conduit builds, with little to no hope of actually making something viable. Either you have to make it a conversion set, or you need to create so many MIs supporting that conversion that they might as well be in a set, cause they’ll only ever be used together.

And as Protoss said, yeah, people will post more feedback about sets so sets will get more balancing. It’s logical. If you post a custom Fire FoI build and complain that it’s underpowered, within 60 minutes you’ll have 4 people picking apart your setup and suggesting you try that MI and that MI instead or that you really should be using a different pair of pants or whatever. Every single aspect of that build will be questioned, because with more customisation come more points to potentially criticise, and since there is no “one true undisputable setup” for greens, the devs will be more reluctant to balance around your setup when there are half a dozen other setups that may possibly perform better. It’s just too fucking time consuming to test them all. If a set-based build sucks, it’s much easier to attribute it to the weakness of the set, because it takes up such a big portion of the inventory and is thus responsible for a lot of those weaknesses. It was no wonder Ultos was/is hell to build around when it provided the second worst resistance lineup in all 5pc sets. Pretty much the only thing you can balance in custom setups is damage output, because survivability can be so massively variable based entirely on your affix rolls. So you’ll have, say DEE MIs quietly getting some extra damage or RatA or lifesteal patch after patch with little fanfare cause they’re just one piece of a bigger puzzle.

TLDR: all design variants of sets are fine and have their place. Not everything should be transformative, else the game loses its foundation. It’s true sets are getting a lot of the attention, but that’s kinda logical. The one thing I’d keep an eye on is when sets for a specific skill are getting buffed, are the MIs getting buffed similarly alongside that so that they retain their relative power. Not all burden can be put on devs, if the community doesn’t provide feedback for custom builds for fear of getting their setup criticised on every aspect, relevant or not. I’m sure if I had raised my voice about custom Mortars getting punted into obscurity by Pyran CDR buffs, Zantai would do something about it.

4 Likes

there is no point to test something so obvious at the moment: Demonslayer got stripped of extra projectiles and Chaos/Cold PB conversion while custom arbalest will have those conversions AND extra projectiles. I compared tooltips of full demonslayer + green oh versus three piece Demonslayer + Spectral Arbalest just before this latest iteration ot test patch and tooltip was identical. And after the latest quadrupple nerf full set PB’s tooltip went down by a lot so I don’t think you need to do the field comparison tests to conclude that current Demonslayer is completely butchered (and I also did test current full set obviously).

To be on topic I agree with most points except I am a sucker for glorified recoloring - play same skill in different flavours work for me. Mainly because different damage type not just look different they feel different too and for example builds around Vitality PB, Pierce PB and Aether PB will look and feel different.

3 Likes

I approve of the sentiment.

People jumped on the “glorified recoloring” wording but the meaning behind it is valid. Yes, recoloring is cool and helps with the build’s visual identity when you play. But when all the set does is recolor, it brings up some dystopian illusion of choice vibes.

I think with the new FoA tech it will be possible to make more sets do smth unique to the supported skills. But it’s gonna take a lot of work to implement and balance out.

1 Like

I was thinking exactly this, and its backwards compatibility we’ve already seen it can have with sets like the recently addressed Radaggan set. In the end, whatever it’s done will need a lot of work, but from what I’ve seen, at least Crate is on the right path.

i think set designs are largely fine, even if not perfect or even uniform.
I love “recoloring”, lord knows i made conversion memes before they got officially enables by sets or conduits just for the heck of it/because funny and entertaining.
And while i think it’s a great notion that sets would make things more special, ala mentioned conduit effects/aether PB mechanic change, it’s also a double edged sword depending on what that change or new effect is, which leads into the next.
What i don’t like, or rather what i don’t think is as good, is when sets end up being subpar or even avg. What is the “point” of being forced into 4-5slots, when you can just do individual items or “custom” setups to achieve the same or better. Recolouring or just generic dmg boosters really being the norm there, but with potential mechanic/effect change it might be worse, imagine current statted version of chaos Trozan etc being gated of to 4p voidsoul or something, would you really want to sacrifice so much for that?

Atm we had a different conduit meme being moved to set, Acid RoK/Radaggan, which i think can be fine, but with conduits there was an underlying/“inherent” understanding you should expect it to suck, so now acid RoK must be good or it will be a waste to build for on 4p set. (this is not a critique that acid RoK is/will suck just point that it can’t be “allowed” to the same now/by default)

Another example on the generic front might be the recent fall from grace “ever” lambasted Rah’zin, which does nothing better than a lvl 55 component, yet locks you into 4 items to get comparable or less chaos build dmg as piecemeal or another chaos set, “what’s the point of going rahzin other than equipping the (cosmetic?) Rah’zin title on your char” it could be argued(was actually argued :sweat_smile:)

TLDR i don’t think it’s an inherent flaw of GD’s design approach, but limited by balancing concerns and performance equality that we “must have”.
Sets should be powerful and cool, but if they become too powerful they become even more standard/all other items/modifiers drop in appeal.
And if you gate everything special off to sets too/effect changes, you run into that issue of reducing availability and quantity that might otherwise be easily afforded on 9x conduits etc.

I think the best way to handle it might be going “halfsies”, accepting some sets just arent’ stellar, while others are, applying a sorta conduit tolerance too sets. But then also adding that proper conduit “umph”/significant change to some sets, so if it’s in the “Rahzin range” atlest it might feel special even if not performance standard standing out.
Current Radaggan change might be a decent example of a switch-up leaning in the right direction, mediocre to more special (tho i can’t say anything for its meta performance direction/impact) - also even if in the end it’s in essence just a mere recoloring set/change still :smile:

2 Likes

Makes me wonder how it is possible to add Kraken for 2H arbalest while maintaining Rattosh and Abomination with Bat, Scales and Hydra. Sorry for offtop,

Rough sketch

1 Like

you can do pretty much most normal variations fine with just Rattosh+Abom or Rattosh+DG
the issue you run into is when you want all 3 Level 100 (GD 1.2.1.3) - Grim Dawn Build Calculator because it’s so tight on points there is only room for minor tweaks, and nothing that costly as hydra or flat RR ( think there was a version squeezing in kraken but still not RR let alone hydra at the same time)