Gem Rebalance (Not Legendaries)

I’m going to talk about several different aspects of Gems here. Just so you don’t feel like you have to read all of it, I’ll put the important information in Bold, to skip all of the explaining and get straight to the point. To avoid making my posts longer than they already are, I only mention the stats for the Flawless Royal gems; obviously the previous levels would also be rebalanced accordingly.

The first thing I’ll talk about is Weapon Affixes. The only time you see anyone use something besides an Emerald in their weapon is when they aren’t intending to do damage with their attacks and damage-based abilities at all. The other gems’ bonuses aren’t as useful for their own tasks as the Emerald is for it’s task. They get used when they’re the only thing that will help. They should be buffed up to the point where they’re actually options that are weighable against the other options, rather than just something you pick because it’s the only thing that helps.

We’ll start with a Ruby, because it’s directly competing with the Emerald as a gem for increasing the player’s damage output. That’s a tough spot to compete in. If a player has built up the 55-75% critical hit chance that most players have built up, then obviously the emerald will be the best choice, as it should be. But if the ruby’s going to have a time in the game (besides leveling heroes to 70) where it’s useful, it should at least manage to be useful for players who haven’t built up so much critical hit chance and damage yet.

At 35% crit chance with +150% Critical Hit Damage, a Flawless Royal Emerald is kicking that up to +280%. Crits increase the DPS by 98% instead of 52%. 198/152 is 1.30, your total damage output has gone up by 30% of its previous value despite not having a lot of crit stacked. A ruby is adding +270 Damage. Let’s say your weapon hits fast and does low damage. That’s the situation for the highest percentage-based increase in damage output you’ll get from a Ruby, so if it’s not useful there, then where is it useful? So you’ve got a decent non-ancient Legendary weapon with 2380 DPS, and its average damage per hit is 1700. If the Ruby is adding 270 DMG to a weapon with mere 2380DPS weapon with 1700 damage per hit, which is highly favorable conditions for the Ruby to have a bigger impact, then the Ruby adding 270 Damage will increase the DPS up to 2760. That’s a 16% DPS increase. Even for a player with sub-par gear where the Ruby is more useful than it is later, the Emerald is still twice as effective.

I wasn’t originally going to suggest increasing the Ruby to the point where it could ever compete with Emeralds because I didn’t think the player base here could handle that, but I will suggest increasing it from +270 up to +600. Even then most decent builds will gain more from an emerald than a ruby, but it might find its place in a build that can’t crit that was previously impaired significantly by rubies being too far inferior to emeralds. It will also finally be slightly better than an emerald in extremely fast-attacking weapons that are used with an offhand; but if the player is dual-wielding, the emerald which buffs both at once will still reign supreme.

Next, the Diamond. Increases damage against elites by 20%. What good is +20% DMG to elites when an Emerald can easily make a bigger difference than that against everything including elites? If the Diamond is going to have a purpose, it needs make players more effective against elites than the Emerald, since they’ve just sacrificed a vast amount of damage against everything else. In most damage builds, an Emerald is increasing their Critical Hit Damage from 300-350% up to 430-480%. Let’s say 325% up to 455%. That’s bringing your Crit output up by 40%, and depending on your Crit Chance your total DPS is probably going up by 25-35%. Let’s say 28% for 60% CC. If an Emerald is giving you that much damage all the time, what good is a Diamond that only gives 20% sometimes? The Flawless Royal Diamond should give +40% Damage Against Elites. Even then, when the player uses a Diamond instead of an Emerald the player is giving up 28% damage against everything else to gain 12% more damage against elites. That’s not overpowered, it’s just beginning to be useful for specific situations. You might see some players start throwing a Diamond in one of their weapons when they’re dual-wielding, or both if they’re a less progressed player fighting Infernal Machine bosses who only needs damage against Elites.

Next, the Amethyst. 28,000 Life Per Hit. Not bad at all. If you want your weapon to increase your survivability, the Amethyst will do that job, and it’s very good at it. You don’t see it used often, but I don’t think it needs a buff. Dual-wielding with Flawless Royal Amethysts in your weapons will give you unparalleled healing. I won’t suggest buffing it until people use it, it’s already fine.

Topaz: +38,000 Thorn Damage. This is a tough one. A player really has to do a lot of building to get their thorns to be useful, but once they do, 38,000 Thorn Damage is a lot. Until that point, it’s totally useless. I don’t think there’s a good solution for that problem. Thorn builds have finally become a usable option, and there’s even more than one that works. The balance may not be perfect, but I don’t want to mess it up trying to make it better, so I will leave the Topaz alone for now.

Old Gems on Weapons:
Amethyst: +28,000 LOH
Emerald: +130% CHD (Typically 25-35% DPS against Everything)
Diamond: +20% Damage Against Elites
Ruby: +270 DMG (Typically 7-18% DMG in a Weapon)
Topaz: 38,000 Thorn Damage

New Gems in Weapons:
Amethyst: +28,000 LOH
Emerald: +130% CHD (Typically 25-35% DPS against Everything)
Diamond: +40% Damage Against Elites
Ruby: +600 DMG (15-35% DMG in a Weapon)
Topaz: 38,000 Thorn Damage

=====

Offhands getting one socket to put a gem in for a mere stat bonus is very underwhelming. There are so many useful affixes that can go on offhands. And then the socket just gives one normal socket? If it’s going to just give primary stats, then it should be able to roll two sockets. That still won’t make sockets a very desirably item on offhands, but at least it won’t feel like a complete slap to the face. Or, if you want a more interesting solution, I would like to suggest Offhand affixes for gems. This one isn’t easy. Here are my suggestions:

Plan A: Increase socket capacity of Offhands to 2.

Plan B: Add the following Offhand properties to gems (but don’t give them two sockets):
Amethyst: +10% Healing from all sources
Ruby: +8% Damage
Emerald: +7% Attack Speed
Diamond: -12% Damage Taken from Elites
Topaz: +10% to highest static Elemental Skill Damage (If tied, bonus is split between them)

This would make the socket a useful but not mandatory affix on offhands. On shields, which don’t have as many good affixes, it would even become a preferred affix to have. Some players may look at this as “Another affix we have to get on items, making it harder to get good ones”. The opposite is true, more affixes on an item to be desired means better odds that you get desired affixes on the item.

=====

The gems in Helms need some rebalancing. There just aren’t many valid choices for players.

I’ll start with the Ruby because it needs help the most. It currently gives bonus experience, but that bonus is reduced by 90% at level 70. Considering that players won’t have good quality rubies until they hit level 70, this isn’t helping a whole lot. If they are trying to level a new character afterwards, there are already a ton of options they can use to get that character leveled up faster. The ruby already has a tough time being useful as it is. It got the short end of the stick everywhere. About the only thing it does get used for is Strength for a Barbarian or Crusader in his armor. I get that you don’t want players to be leeching on public games with a gem in their helmet. You don’t want the highest level players to just be people who had teams that allowed them to backpack along for EXP with a ruby in their helmet. Here’s the solution I suggest: Let Rubies give their full exp bonus at all levels. Instead of reducing the bonus by 90% once they reach level 70, just reduce the bonus by 90% while they are in a party with players. Now the Ruby becomes a solo leveling tool. A player using a ruby in solo will have given up their helmet socket, the team bonuses, and the team to clear rifts faster, but at least they can try and make up for all of that with a +41% EXP gem in their helmet. Solo players have been wanting something to help them keep up; I think this is it.

The Diamond is already useful for some support players. 12.5% Cooldown Reduction is highly sought after in some builds. It’s fine.
The Amethyst easily proves its usefulness with +23% Life. That unparalleled bonus to toughness needs no buff, but it’s not overpowered either.

The Emerald’s +41% Gold Find is unecessary to most experienced players, but if you’re just speedrunning anything other than Greater Rifts, it can be nice to also get more gold while you’re doing it so you don’t have to save those dang Puzzle Rings for gold runs. With Gold Find being multiplicative and also helping get more gold from Boon of the Hoarder, it’s a useful thing to have. Finding more gold means not having to spend as much time with gold farming equipment on. Still, with Gold Find being so abundant, I think this should be increased to +50% Gold Find, allowing the right helmet to kick that up to +100%.

Next we have the Topaz. This reduces resource costs by 12.5%. With resource costs already being very well-managed by most players, this isn’t a very useful bonus. Resource cost reduction is definitely less valuable than Cooldown Reduction. I suggest that this be increased to 15% Resource Cost reduction, which would allow it to at least equate to 15% damage reduction for the Crimson set, making it a bit more comparable to the Amethyst for survivability at least in that specific instance.

New Helm Gems:
Ruby: +41% Experience (4.1% in a Party)
Diamond: +12.5% Cooldown Reduction
Amethyst: +23% Life
Emerald: +41% Gold
Topaz: +15% Resource Cost Reduction

=====

You probably thought I was done talking about Gems. If I’m not talking about Legendary Gems, what more could I possibly say? Go on to the next post if you want to know.

Belts. What better place to put a gem than on a belt? Yeah, I’m about to suggest adding sockets and effects to belts. If that’s too much for you, I get it, but here’s my idea for what gems could do in belts:

Belt Gems:
Amethyst: +30,000 Life Per Kill I am aware that items normally roll 6,000 LPK. That’s a secondary affix. This is a gem put into a primary affix that only holds one socket. How does that LPK stack up against Life Per Hit?
Ruby: Chance to deal +20% Area Damage
Emerald: +5 to Gold/Health Pickup Radius (This value just won’t go up with every gem level, it would go up every few levels until reaching 5)
Diamond: Attackers have a 50% chance to be stunned for 1 second. (Lower levels have lower chance)
Topaz: Enemies are slowed by 80% for 5 seconds the first time they come within 10 yards (starting at a lesser slow and increasing to 80% with each gem level).

Yes, I did think about how each of these values could be utilized by various builds that convert previously useless bonuses into useful effects. They still won’t be overpowered, but they will become desirable yet not mandatory bonuses for those builds.

22 Likes

Good solid post mate, well done.

2 Likes

I think not every gem has to be ultimately useful at the end game. Some gem sockets are meant for followers (ruby or diamond on their weapons), farming (emerald on broken crown with goldwrap) or leveling up phases(ruby on helm).
And if you want “balance”, you have to mind legendary powers of leoric’s and broken crown while doing a design about it. As their power and drop rates can be easily manipulated.

A ruby or diamond doesn’t really help the follower. What good is more damage when they won’t kill anything on the difficulties that you actually play? It can be fun to watch them kill things on Normal but that’s about all you’d get out of them. If you really try hard, maybe you get your follower able to kill Torment 1 with patience.

Getting extra gems doesn’t really affect the effectiveness of those gems, so I don’t think that’s relevant here.

As for doubling the gem’s effectiveness, you’re dedicating a legendary affix to that. That’s the result of that legendary affix, and not the gem itself. Your helmet or kanai cube could be giving a different effect instead that would usually be much more useful than doubling the effect of a gem. Either way, to count the doubled effect of the gem as though that were the effect of the gem itself when considering balance would be asking for severely underpowered gems.

I did not suggest making every gem always good at endgame. If you read my post, you’d see that for some of them I was just suggesting at least getting them to the point where they’re useful for players along the way.

Yup, I skimmed through it at first and appears I missed all the good details. My bad.

It supposed to help when you’re around T1-4 just to get health globe from elite/champs faster. When you reach the power to farm T16 their dps contribution is minuscule and insignificant as you clear an entire screen with flick of a wrist.

1 Like

Yeah, I geared my follower for damage output before just to watch them do T1. It was cool to watch. But it’s just not worth it, their real purpose is to support you. Throw crowd-control on them along with anything that makes enemies take more damage, and get them as much attack speed and cooldown reduction as you can. Any procs that hit more enemies which will trigger other debuffs or crowd-controls are also useful.

Really cool post and ideas dude. Ruby in a belt would be damn cool.

What I would say that I don’t agree with though is that the rubies aren’t bad when it comes to leveling even though you only get the lower quality ones for most of the process. Getting even a Flawless early on is very useful, 15% experience is nothing to scoff at. I often get Flawless Squares, 25% is a big reduction in leveling time. And with getting Marquise from 60 and not earlier, the last 10 levels take the longest and 33% is damn big. In short, the fact that you get lower quality gems for most of the leveling process doesn’t make them bad since the XP increase is a big deal even with the 15%.

Also rubies are used for more than XP/Damage in leveling and early game. You use them on Intelligence classes on armour just like emeralds. I assume you know that and meant something more in the direction of “we have emeralds for that”.

I am also fine with not all gems being used for end game, as long as they are useful at some point. Yes the ideal is for every gem to be useful in every slot but if for example a ruby is only useful during leveling and then for Int classes I think it is OK or an emerald being mostly useless in helm but good in the rest it is also fine. But good job on trying to get to the ideal.

My suggestions weren’t tailored to make every gem useful for end-game. I even specifically mentioned that. But I do think every gem should still be useful at level 70. I don’t consider level 70 to be endgame. Any gem that becomes useless by the time you hit 70 is useless after 1-5% of your playtime. I think they should at least manage to remain useful until you have good gear.

1 Like

Think I used some bad wording… Didn’t think you meant it that way, said it more as a general point.

It is a good point yes.

You didn’t use bad wording, it’s just that on my end I assumed what you said was intended as a reply to my post. I get now that you were just saying your opinion on gems, not replying directly to mine.

1 Like

I was agreeing with you up until this reply… I’ve gotten my merc to have decent throughput if you stack enough elite damage, It’s been a 6-7 seasons since I’ve played seriously but I remember seeing a prettyyy high crit on an elite, had furnace augmented with diamond/soj/unity all augmented and every item with the highest values… Was it game changing? no but having a merc hit for 300-500mill+ on elites is nothing to scoff at. Don’t get me started on how valueable the aura that bulkatho’s provides, all irrelevant though when you get to a certain point… Heck i’ve been hoping since they implemented legendary gems that they’d make a handful for mercs… I digress, these are some of the problems i’ve always thought diablo 3 had, gems and merc power…

I agree with you completely on the lack of agency/choice behind the gem structure, they really should make ruby compete at least mathematically it should have better sweet spot for usage. They probably should have just limited the amount of crit hit damage long ago so the choice would be more clear --> once maxed crit hit damage then you could use weapons for survival/elite dmg or raw or prioritize cdr on all jewelry and use weapon to boost crit hit dmg.

I like the belt/offhand idea, gives you something to think about really… The devs have been good to us lately with the changes so maybe they’ll read this and consider implementing some more options or even some specific situations where you can double socket offhand/weapons with specific rules like 'cannot use the same gem twice or --> no emerald if you use Ramaladni’s Gift on it twice, etc.

Sorry, but even 500 million damage from your follower, though impressive for a follower, is not contributing. I can hit for 10-30 trillion on rapid-attacking abilities that hit multiple targets. Your follower would quite literally take hours to deal that. Granted, that’s with buffs, but even without temporary buffs I can reliably hit for 3-10 trillion. Or I can use a strafe build that sends out an absolutely absurd amount of projectiles that deal around 500 billion each. Like I said before, you can build your follower’s damage up to some nice levels for fun, but their damage dealt is just not going to keep up enough to help on any difficulty where you actually need your follower’s help. Maybe you haven’t seen what the set bonuses are like nowadays.

Legendary gems don’t work on followers, unfortunately. Ice and Poison on them would have been great.

I like most of your ideas. Gems in belts would be cool.

Some builds desperately need the topaz in the helm. These builds usually want CDR also. As such, I’d prefer diamond or topaz in helm doing both CDR and RCR, and the other one doing something else interesting.

That’s hard though, because emerald in helm is already useless, so I’m not sure Blizzard would be able to come up with anything else interesting. Bane of the Hoarder is all that’s needed to get more gold than you need.

yeah if you took the time to read my post, that’s my gripe is that they never made a legendary gem to specifically work on mercs --> hence my rant which was also to point out how it’d be nice to have some of these extra utility options to buff mercs. Also your whole point is moot because of how scaling works in greater rifts, it doesn’t apply to mercs because the ceiling is much much lower. When I was playing 500 mill attacking enchantress could have been the difference in completing that rift in the last second or two and placing on leader boards. I’d play d3 more often if there were more agency with gems AND mercs, which was the point of my entire post… Sure the power creep is so high past 60ish solo that it makes 500 mill dmg completely irrelevant but that’s my qualms with the game --> gems don’t scale well neither do merc’s. If they did I would be playing more and hitting higher greater rifts to augment, in turn I could probably turn that merc damage into a billion easily, especially with some of the gem changes you are proposing. Which is why I’m advocating for these changes but I just have a different philosophy --> merc’s should be viable outside of utility, even if its just .001% of the bosses hp per attack.

I agree that mercs should be capable of going for damage builds. Letting Legendary Gems work on them would be a good way to fix that. I honestly don’t think letting them use the currently existing ones would be that big of a problem for balance.

I did read your post, I saw what you said about legendary gems not working on mercs and wanting merc gems. I was just commenting that unfortunately we can’t do that. There was talk of considering merc gems years ago but that seems to be long forgotten.

The ruby on weapon suggestion doesn’t make any sense.
People would still use emerald for more damage, and if the value is made higher to compete with emerald toe-to-toe, then it still won’t make sense. You’d just have two gems that do the same thing - it’s just that one increases one number, another increases different number.

The only build that doesn’t use critical damage is thorns, and thorns doesn’t use weapon damage at all.


The problem here is in scaling of gems with items they are socketed into. This is one of the main reasons, why we have “good gems” and “useless gems”.

You brought a good point earlier in your post, but you didn’t expand on it - fast weapons and slow weapons. It would make sense to increase critical damage of slow and hard hitting weapons, but increase the base damage of fast low damage weapons. This, however, requires a few new mechanics implemented into the game:

  • Most skills would need to depend on your attack speed of the weapon. You can’t have a hard hitting weapon equipped, but your skill attacks at lightning speed, so you don’t lose anything.
  • Again, scaling for different sockets.

For instance, let’s look at a 2h crossbow and 1handed ones right now. If you want highest consistent damage, you’d run 2h crossbow + quiver. You get highest base damage, a decent amount of CHD and high CHC, and your attack speed doesn’t really matter much. In groups, you’d go for double 1h for extra CHD, because your support gives you CHC.

So, using this example, there’s no point in which you’d use a ruby here.

Now, let’s modify the Ruby AND Emerald:

  • Ruby - gives more damage the faster the weapon is. Thus making AS useful affix on 1h weapons. (low bonus on 2h)
  • Emerald - gives more CHD the slower the weapon is. This makes 2h weapons good for nuking. (low bonus on 1h)

This redefines the purpose of these weapons and the builds they are used for. It promotes diverse playstyles and builds. Attack speed builds would be useful now.

This is just one example for change of gem for one type of socket. This kind of improvement can be made across all different type of sockets to improve overall customization of characters and allow for various different types of builds and playstyles.

Don’t just try to make things as useful as the best thing by just continuously increasing numbers; redefine their purpose and introduce new use cases.

More use cases = more flavorful and more useful product. (I think this can be said pretty much about a lot of things in life and business.)

There are some abilities and item affixes that don’t crit. If the Ruby were buffed up to a somewhat usable level, you might see someone design a build focused around pure Damage. Yeah, my buff doesn’t bring it up to par with the Emerald, because I don’t think the developers ever intended anything to compete with the Emerald. A buff to the Ruby that actually made it a good gem would be met with outrage from the community.

+400 Damage from the Ruby can add over 600 DPS to a fast-attacking 1-hander. That can be enough to make it almost as effective as an Emerald, except it doesn’t require Crits to be useful.

You might not think the buff is enough, but we should be able to agree that it’s a step in the right direction. Would you really prefer that it stay at 270 instead? If what you really want is MORE, then maybe you should be pushing behind me instead of against me.

You are correct. Nobody uses any builds that don’t crit because abilities and affixes that don’t crit become useless, but they do exist. I don’t expect a buff to the ruby to be met with a massive rise in no-crit builds. I still don’t think it would be used much at all. But at least newly level 70 players with fast-attacking weapons who haven’t stacked up much crit chance on their gear yet might be able to use a ruby for an hour or two. That’s about the best I can ask for. I anticipate a lot of backlash from the community if the Ruby were buffed up to usable levels. For whatever reason, they consider +Damage a “flat, stupid, boring affix” while seeing Critical Damage as the “sophisticated man’s damage” as if they had to aim for the Critical Hit themselves. They don’t want +Damage becoming a sought-after bonus.

Only the ruby directly affects the item it’s socketed into, adding flat damage and thus making it more useful on faster weapons. The rest go onto your total stats, just like everything else does, and are unaffected by the stats of the item they’re put into. We have “good gems” and “useless gems” because their effect is good enough or not good enough, not because of how they socket into the item.

This doesn’t require any new mechanics added to the game. It already works, it just needs an adjustment.

The attack speed of abilities already adjusts based on the attack speed of your weapon. Faster abilities like Strafe and Whirlwind might still be fast, but they are noticeably slower with a slow-attacking 2-hander versus dual-wielding fast weapons. If the ability’s attack speed didn’t adjust, you would just see everybody running 2-H whirlwind builds.

I don’t know what issue you’re pointing out by this. Please explain.

You’re right, there is no point in using a Ruby for a Demon Hunter. Unless you were running a Vault build with Trail of Cinders, which does not Critical Hit. Sadly, that build became garbage long ago. Maybe it will become useful again some day.

With the bonus Rubies give going to Damage, not DPS, it already adds more DPS the faster the weapon is. A weapon with a 1.6 attack speed will receive 60% more DPS than one with a 1.00 Attack Speed. I don’t think this difference needs to be made even more extreme.

This would essentially just be a mega buff to 2H weapons and a mega nerf for 1H. It would throw off the balance a lot, and it would take a lot of work to restore balance afterwards. I don’t like this idea either.

They already are. Attack speed is an extremely valuable affix. Just about the only thing limiting it is that its values are usually lower than comparable Damage or Crit bonuses, but even then it’s still highly sought after by some builds, until they’ve obtained their desired attack speed.

This is exactly why I don’t think forcing Rubies onto 1H weapons and making Emeralds useless for 1H is a good idea. It means less choices, not more. Buffing the Ruby up to a useful level in a way that doesn’t pigeon-hole it is what would give players more choices, because then they would actually have to decide for themselves whether a Ruby or Emerald will be more beneficial to them.

I might have disagreed with everything you said, but please do not think that I am attacking you. I appreciate you posting here and discussing ideas with me in detail. If we agreed on everything, there wouldn’t be anything to say.

Invoker says hi.

:rofl:
Sorry, I meant Damage-based builds that don’t crit.

Balance restored?

There’s not a single solo build that uses AS on rings, weapons, and gloves for DH.

And that’s the point. Why do AS break points exist? Because overall, AS is just useless compared to CHD+CHC. You should have an option to just ignore critical damage and go for maximum AS you can get without any limitations. You can get as much CHD and CHC as you can, and your damage still improves. Why can’t AS be the same?

It doesn’t limit the choices. Your 1h attacks faster, so if you have the same amount of CHD as in 2h, what would be the point of using a 2h? You could do more attacks that would have better chance of critical strikes and the loss of base damage wouldn’t matter in that case.
Bottom line, you’re not forcing rubies in 1h. Because remember, in order to use rubies in 1h efficiently, you give up on your CHD, CHC on your other pieces of equipment for AS to maximize the value that your 1h weapon gives you.
You could still go for emeralds in your 1h, if you so desire. But the takeaway here is that you not only give up CHC from offhand, but also a bit of base damage, which is compensated by wielding a faster weapon.

However, this is not set in stone. As mentioned in my previous post, this requires new mechanics and possible redesign of some items. For instance, you’d want to repurpose legendary affixes on the weapons to give them better identity.
The attack skills need to ALWAYS work with the attack speed. You can’t have a skill that is inherently faster than another even with same AS values. It shouldn’t work like that.

No, it would just give players another thing to compare with other things that are useful “across-the-board”. You need every item in a game to have some form of identity. (What it is best used for?) People don’t run Avarice in GRs because it is useless there. People don’t run Stricken in easy content because it’s useless there. People don’t run Ruby in weapon because Emerald is better there. You see the difference?

And in regards to making it easier to level characters…
Are you out of your mind? Leveling is easy as it is, even for a completely new character with no help. We already use rubies for low level because we don’t have CHC. So, we kill something a bit faster, so what? In a few hours we’ll go back to using Emeralds and just forget Rubies ever existed.

Once again, this is just ONE example – making Ruby’s number higher a little bit is pointless. If you want to do that, then do it in a manner that it would compete with Emerald (which is still a terrible idea, but easiest to do).
If you want to bring diversity, choice and balance, you need to go back to the roots of RPG and bring back elements such as “item identity” and their “use cases”.
Increasing numbers is just lazy and completely pointless (if not increased by a substantial amount).