Arena Competition is once again Dead - Snapshotting needs to be removed

General
1

Let me preface this with that I have been playing on and off for around 5 years waiting for a competitive platform in this game for what I enjoy - competing on the arena ladder against the best players, and it hasn’t arrived, nor do I think it will ever arrive with their current fundamentally flawed approach to balance.

The current situation? We have a forge guard manifest armour build at the top of the arena ladder nearing 2k waves, which swaps out its entire gearset/weapon/idols to T7 Phys dmg over time / increased minion health /damage / regen / DoT, the whole works, to snapshot stats and summon a manifest armour minion that is a titan of damage and survivability, then swaps back to its entire defensive focused gearset of 6k+ life, 2k+ hp regen, 100% block with 4000+ block effectiveness, capped armour and everything else.

The result is that the manifest armour minion runs around destroying everything because it is a titan of stats that should never be possible, while the player is a titan of a defensive tank that would never be possible on any other build because DPS would grind to a halt and not be able to damage anything, not even white mobs.

This is all possible because of snapshotting, which should have been removed from the game years ago, but is still here past 5years of EA and 1+year post release.

There are no other builds that can match the defensive and offensive power that snapshotting gives. Any other build that tried to match it defensively would have no space for damage; any other build that tried to match it offensively would be a glass cannon that falls over as soon as a few white mobs looked in its general direction.

There are only a few other possible viable build options for arena this season, and they too abuse snapshotting: Abomination necro. Once again, you swap out your entire gearset and idols to snapshot and summon a titan of an abomination minion that has a ridiculous amount of damage and survivability, then you swap to your entirely defensive focused gearset of defenses, damage mitgation and 100% block then hide around a corner while your minion runs around smashing things for you.

And maybe a few outlier builds that I find are completely cringe and shouldn’t exist, like the ballista build that has -200% resistances and completely ignores the defensive aspect of the game because its entire focus of the build is to 1shot the entire screen from over a screen away.

Other than that, there are no other contenders for competition. We have another dead season for the competitive platform of the game unless you want to play one of the snorefest’s listed above.

Part of why this is a problem is because of snapshotting (which is basically a form of bug abuse). The other major part of it is the approach to balance, which I believe is fundamentally wrong. A reply to a post I made on this exact topic months ago basically sums it up my thoughts:

We still have a balancing approach which only focuses on nerfing the outliers / top end of builds, and completely ignores the rest of the population.
I DON’T have a problem with that. Overperforming builds probably do need some adjustments to bring them in line.

But where are the buffs for all the underplayed and underrepresented skills? The least they could do would be adjustments from patch to patch to bump up the base damage or effectiveness of some of the skills that are never used.

Some guy in global chat was enjoying acid flask, and asked if it could be made into a build in the endgame. I felt bad telling him that there was basically no chance because there was almost no scaling in the skill, its basically a dead end, and probably always will be.

Some other skills that come to mind? Puncture.

Glacier sorcerer, ever since they completely killed off that skill by making a targeted glacier have ZERO crit multiplier, effectively making it a dead end and eliminating what was a pretty fun build from the game. This was a popular and fun build that was played widely until this change was introduced, now it has a 0.1% playrate when compared with other skills. It is functionally impossible to build because of this one limitation, locking people out of the only way it was FUN to play with: when the skill was targeted.

Earthquake werebear: the build was basically deleted from existence with the heavy handed change that made it so the item enabling the build had a limitation that it could only deal its main damage once every 5 seconds. Imagine if any other build you played had to wait for 5 seconds dealing zero damage until it could deal damage again. You would quit the build out of fustration or boredom. Yet another skill which is now, 0.2%-0% playrate.

Upheaval could have had some cool builds. Too bad the scaling and base damage is non existent. Another skill that is almost never played, has almost no builds for it and yet still received zero changes in previous patches to bring it up.

Volcanic orb? Cinder strike? Poison flask? More skills that are basically ignored patch to patch even though they have non existent playrates because the base damage or scaling is non existent, its not possible to create a build from them without experiencing constant fustration or inconvenience to the point it simply isn’t fun to play anymore. Yet they got no attention or changes for several patches so far.

And now these recent patch notes, one major global change / sweeping nerf I saw was changing the ward retention per point of intelligence from 4% to 2%. Another change that is targeted towards the topend ‘outliers’ and way too heavy handed. That is a huge sweeping change that essentially cut the ward retention (essentially lifepool) of an entire archeotype of builds by half. This is in combination with Twisted Heart being nerfed 3-4 times in a row and now only provides 1/3rd of the ward generation that it originally had. Now we have essentially what is a ‘health meta’ because of it.

Such massive changes like that should have been made small and incrementally. If anything, it should have been a change from 4% to 3% first.

You know what patch notes are boring to me? Patch notes that simply go over all the skills that we know are good/fun/strong, that lists many changes that incrementally make them less fun and more inconvenient to play.

Patch notes that I could get excited about? Where they notice skills that are underplayed and underrepresented and make meaningful changes to them from patch to patch, so we can see some different builds and possibilities to theory craft in the meta, for more diverse build possibilities.

But I don’t see any focus on balance being made towards a diverse build variety. Just the same ‘we are pulling back builds that are overperforming as outliers’ changes, and pretty much ignoring every other skills in the game. Every patch. For years.

And it results in this, an endgame arena where we have zero build diversity or options to compete if we did want to compete against the other best players in the game. Because competing would mean playing the exact same god damn boring build.

I know that not everybody cares about this, but I am pretty sure that there are a few other players out there like me who would like to compete against the other best builds/players competitively. The reality is that there is no competitive platform to do it right now, because of this state of build balance, it is non existent. There is not a single build out there that can match the ridiculous defensive or offensive capabilities of the current powerhouse that snapshotting creates, it simply is something that should not be in the game at all in the first place. And its such a simple fix to get rid of it- make the summons be resummoned every time they change instances every 5 waves.

Other than the obvious snapshotting issue, we have a whole plethora of builds that simply don’t exist- Glacier being a prime example of that.

I got myself to the endgame and finished my Tier-2 blessings and after some thought, I realized there was no more reason to continue playing or pursuing the objective that I had in my mind the entire time- which was to eventually compete on the arena ladder. So for a player like me, who lives for the arena ladder, there is essentially nothing left for me to do.

I just wanted to come and say my bit on here and see what other people thought about it. I still believe their approach to balance, which is a focus on nerfing ‘op builds’, with zero focus on buffing ‘underpowered builds’, will never result in a healthy, exciting, diverse build meta where it is fun to theorycraft new ideas. It simply can’t and never will reach that state. And because it can’t, we will never see a competitive platform on arena or anything else in the endgame.

And I think that is ultimately a big shame.

2

I just want to reply to this point to say I heavily disagree with this. Every season, they buff a mastery or two. Last season was FG, this one was Pally and VK, next season will be Lich (and possibly Necro).
They don’t buff by skills, they buff by mastery. Because things are always interlocked and it’s easier to fix the whole than apply band-aids.

2 Likes
3

If they were buffing by mastery, then the skills of those masteries should be included.

Glacier playrate has been non existent since the introduction of the zero crit multiplier if the skill is targeted. Why would anyone want to play a build where its impossible to scale your damage because you can’t crit?

Earthquake werebear became non existent when the item enabling it gained a 5 second cooldown. Who would want to play a build where they had to stand around for 5 seconds every 5 seconds doing nothing and waiting to be able to deal damage again? People would quit out of fustration, inconvenience and boredom. The EQ werebear archeotype simply no longer exists anymore.

Poison flask base damage, damage effectiveness and lack of more multipliers means it never has and probably never will have any builds situated around the skill.

Upheaval fundamentally feels good to play. Too bad the base damage is 2, and has very little scaling on its tree, meaning no builds will ever be made with it, because it can’t do any respectable damage.

Cinder strike has always been underwhelming and I have not seen any significant changes to it in many patches.

The volcanic orb tree is uninspired and boring, and I believe is almost as old as one of the first iterations of the game, yet has received almost no changes in 6 years? Surely they might notice that nobody was playing the skill for 6 years, unless their approach to balance actually does not include looking at underplayed or underrepresented skills. If such a skill never received any meaningful changes in 6 years, is that not indicative or alarming that their process for determining balance changes might have a huge flaw?

Which was the point of my post.

As much as you say they are buffing a mastery or two every season, I see many skills that are not receiving any attention at all, from patch to patch, for years. The major changes are all centered around the heavily played skills or the ‘overperforming’ skills. And most of these changes do nothing except restrict build diversity further, instead of expanding it.

Do I think builds like judgement paladin need toning down? Absolutely, it has access to everything in the game, the mastery provides more power than most other masteries (15% more damage btw), the class passives provides more resistances than every other mastery in the game, the skills it has access to have everything: physical and fire shred, armour shred, applies frailty, vengeance has passive 18% damage reduction, aura gives ailment cleansing, puts down 10 auras on the ground that heal every tick independently for ridiculous amounts of healing every second, several sources of multiplicative damage reduction from different synergistic skills, able to itemize for 100% block, capped armour, access to everything in the game as well as huge amounts of more damage multipliers on its main damage dealing skill. For sure it needs to be toned down and is an outlier.

But changes to those builds should be made delicately as to not be so drastic that they go from widely played to non-existent the very next patch. Which is what has happened to several builds, skills and even entire archetypes in the past, and they never recovered, nor were ever reinspected. Which results in exactly what the opposite of what the goal should be; varied and healthy build diversity, because those builds are essentially removed from the game.

Several of the examples I listed never recovered from the heavy handed changes they received in the past. As such, they now have 0.2%-0% playrate, and if you were to try to make them work, the inconvenience and fustration (because they simply aren’t viable) would outweigh the fun as they simply cannot and will not work. How can it work when your main skill has 0% crit multiplier permanently? How can it work when you have to wait 5 seconds to deal damage once, every 5 seconds?

Those are heavy handed changes that should not have been made because they were too drastic, and resulted in those builds disappearing from existence entirely. So all we get from that is less build diversity in the game overall. Less room for people to theorycraft and be creative.

And the changes that should have happened regularly from patch to patch to bring up skills that have been ignored since forever have never happened. I can go back through patch notes and tell you, they are not changing the values of the base damage of those underrepresented skills. They are not adjusting the damage effectiveness of those skills. The passive skills of those skill’s skill-trees are not getting adjustments or changes like more damage multipliers added to make playing them a possibility.

Those changes are just not there. The focus always has been on the widely played skills, and with heavy handed changes to ‘bring them in line’. Like we see with the intelligence ward retention change: a change that affected not one skill, but a huge portion of builds in the game, to the tune of halving their effective lifepool, because ward was perceived as ‘overperforming’. It wasn’t a subtle change like 4% → 3%. It was 4% → 2%, a massive change because they just took basically away half of every ward build’s EHP.

So now we have a meta where health is superior, a health meta.

I don’t think drastic changes like that are healthy.

If drastic changes are justified for the top end of builds, where are the drastic changes for the bottom end of builds? Is it going to break the game if someone started actually dealing damage with Poison Flask for once?

4

All the skills you mentioned are for masteries that they haven’t reworked yet. So that’s only natural. And since acolyte seems to be next on the list, it will likely take 2-3 more seasons to fix them, at least.

They do add some bandaids to some skills sometimes, but usually they’re not too effective. It’s hard to do so without actually wasting time examining the whole. That’s why they are focusing on masteries. It’s easier to look at the whole picture and then change it all at once.

They’ll eventually reach primalist, marskman, etc. Until then, yeah, it sucks that some skills are underpowered, but it’s not because they don’t care about balance. Quite the contrary. It’s because they do care about it and are trying to do the right thing, rather than just applying bandaids left and right and then having a much harder time doing any real fixes.

Again, that would be because you’d be simply adding bandaids that would all be undone when you reworked the mastery later.
They would have to waste time examining the skill, figuring out a way to bandaid it without major reworks, only to have the same effort months later when they redo the mastery.

And yes, I understand that as an altoholic it sucks to have a bunch of skills that are underpowered. But doing bandaid fixes will only make the task harder down the line.

Balance is hard to get right. It took GGG years to get their under control. And this was mostly because of bandaid fixes. They applied a lot of them, but then added something that would break all the bandaids. So they had to apply more bandaids to the bandaids until the next time things broke.

I don’t expect EHG to get it under control in less than a couple of years either. But at least they seem to be doing it with the right approach.

5

Agree with everything you said about snapshoting… it’s kind of a bug abusing and it should be removed from the game.
That’s a great piece of feedback, which you should probably want to place in the Feedback/suggestions section.

Maybe the second part should be made separate as a different post, again in the feeback/suggestions, because it has nothing to do with the first part. While not entirely true, I get the gist of what you’re saying.
But then, they just reworked the entire Sentinel class, so It’s only natural that the obsolete classes will be played less than the new shiny one. And of course it’s more powerful than the older ones… why would they rework an entire class only to make it weaker?

6

Outside of waiting for mastery reworks, or overhauls of the entire skill trees / passive trees, I really would like to see some bolder numbers adjustments to the weakest skills. Increase base damage, damage effectiveness, the values of each node. They’ve done this bit by bit for some skills like disintegrate, but the changes are still not enough, it’s clear they’re underestimating the impact of their buffs. I’d really like some bigger buffs in terms of the numbers adjustment.

For example, Fireball’s more damage nodes are all 3-7% per point. That’s really puny for an underpeforming skill. Increase it and it might see more play, even if it joins the list of meta builds for a season, why not??

7

Disentegrate is probably going to always be a bad skill. Fundamentally standing still is suicide, and it does nothing special for the tradeoff of standing still.

Yes, this is another one of my gripes, fireball gets only 7% more damage per extra skill tree node, whereas I can see skills like Erasing strike getting 50% or even 100% more damage for some nodes. On top of that, many of fireball’s nodes have massive drawbacks- you get negative cast speed or a less projectile penalty.

I even tried setting up a level 31 fireball on my sorcerer recently, to reach all the extra projectiles along with a cindersong for a total of 6 fireball projectiles instead of the regular 3. I found out that getting that many projectiles actually increased the use time of the skill by a huge amount, resulting in me doing less damage overall even with 200%+ cast speed, because now the skill gets animation locked in the use time shooting the extra projectiles that going up to 6 projectiles is effectively the same as having -100% less cast speed multiplier… Just wow. Really feels like nobody even tested that. Because it sucks completely, and it cannot ever be good, because its essentially broken.

3 Likes
8

@DJSamhein Your initial post is factually incorrect. They buffed mana stacking builds by buffing certain skills (that don’t use mana btw). But have left Meteor in the same unplayable state because EHG refuse to address mana regen. Imagine if health was considered the same way. No leach, no regen, and only channeled healing abilities. Everyone would be saying the game is unplayable that revolves around either ward stacking or potion stacking.

EHG balance team is a joke, and that is objectively true. Just look at the new idols, they straight up doubled/trippled the stats yet haven’t removed or altered the old affixes. Making the old minor idols completely dead items to pick up now. Who balances like that? It’s like I’m playing a damn gatcha game.

9

I just don’t understand why EHG doesn’t implement a check on abilities where if a player changes gear, changes form, or receives a buff, that the game then re-evaluates the scaling of skilled abilities & skills on the skill-bar. Would it really kill the performance? I find it hard to believe that such a system is impossible to implement. At the very least a check on item swapping should be possible without altering performance in a way that matters (I don’t see combat gear swapping as something that should be encouraged for example).
The problem of snapshotting isn’t just with minion builds, it’s also with transformation, and any toggle skill. But minion builds are most affected by snapshotting because it’s easier to pull off and generally has more impact.

10

The anti-snapshot mechanism would be triggered constantly, so if there is any small mistake or oversight in the implementation it would negatively affect pretty much every player in every play session. So even though the fix is simple, the development and testing resources they need to allocate to get it done properly are non-trivial.

11

Really? It entirely depends on how it’s implemented no? The game doesn’t constantly check every millisecond I’m not pressing a skill to activate it. So why doesn’t the game crash constantly due to perma-triggers? Oh yeah, because it only triggers when an action is performed by the player; pressing a skill.
In the same way that skill scaling can be adjusted on a player trigger; swapping gear, or changing forms.
The only thing I’d concede is receiving buffs as triggers, as you can have builds that receive a lot of temporary buffs. Thankfully most of them can’t be used to snapshot (aside from attribute shrines, but that’s an easy fix too, just remove the attributes and buff the shrine effect).
But even when you consider buffs, surely the devs would be able to single out certain problematic buffs for triggering such a check.

And also, are the resources they needed to allocate to make sure the power-crept idols weren’t improperly implemented not a trivial matter? Does it take a whole team to see that doubling the stats of old items (without addressing said items) is probably a bad idea?

12

That is the most likely problem with snapshotting. When you summon your minions, it will check the gear you have and use that. Once it does, it’s an independent “entity” that has no connection to you (meaning that connection is one direction only, for minion count/limit checks).

In fact, all skills (except falconer) snapshot. They are calculated when you use it and not changed afterwards. If you use a skill that has a longer animation, for example, glacier, and you cast one and immediately change gear, the glacier doesn’t change stats mid-way. It will use the initial damage calculation and stick with it.

Personally, I don’t really have an issue with snapshotting, even if I don’t use it. Probably because I’m used to it being around in one form or another since D1.

It even makes sense thematically. You created something. Changes to you after you created that thing don’t affect the thing you created in the past.
It would be like you creating a chair, then losing an arm and suddenly your chair is all wobbly and broken.

I agree that snapshotting in LE is a balance issue. Mostly because you have way too many ways to affect minions that you usually don’t in other games, which lead to extreme results.

But as mentioned, this is a performance issue. Right now the game only checks your stats when you use a skill. But to prevent snapshotting the game would have to check your stats everytime you use a skill, change gear, get a buff/debuff, use a passive, etc.
Not only that, they would have to perform this check for every skill you have specced and in your hotbar, even if you’re not using it actively at the moment.
This would massively increase the number of checks required for every single player.

1 Like
13

Falcon does as well but it’s “re-snapshotted” every so often (how often I don’t know).

14

I seem to remember Mike saying it’s every few seconds (5?), but I can’t be sure.

But the main difference is that the falconer was coded in a different way. You never cast it. It’s always with you. It never dies or needs to be recast/resummoned/ressurected.
So it’s not a minion in the same sense the rest of them are.

You could theoretically apply the same thing to minions, but you’d have to recode how they are cast entirely.
And any way you change it will always lead to a performance hit. All you can do is make it so that the hit is minimal and not noticeable, but it will always require more resources than the simple snapshot mechanic you have now, which only checks once.

1 Like
15

Yeah, it does share some similarities with the Primalist companions.

Yeah & there would be issues with the temporary minions for Acolyte (& Sentinel).

1 Like
16

So why doesn’t EHG change the way skills check and are checked? Would it really impact performance to implement a two way-trigger based check? Are you changing your gear constantly during combat? Why would such a check impact performance to such a degree to make the game worse?
And why is it that falconer can have constant checks that aren’t even trigger-based and not have the game shit a brick? One would think trigger based checks would be better for performance than code that is constantly running.
I also know that all skills “snapshot”, but it’s really not something to be concerned about for most skills. It’s simply an issue with minions and toggle skills.

And just because old games had snapshotting made in the damn 90s where we were playing on rigs with 32mb of RAM. Why is snapshotting still a performance issue? If that was the case, falconer wouldn’t exist in it’s current form.

17

Probably because it would be a massive change that would require a lot of time and effort.

Because that is a single minion that is used in 1/15th of the available masteries. Also because, while it is technically a minion, it can never be targetted. It only targets enemies. So in a way, it works like a regular skill on autocast.

If you consider masteries that have whole zoos, it would increase overall server load for everyone.

Oh no, quite the opposite.
What we have now:
-You cast a skill. It will check your stats and cast it, then forget about it entirely.
Trigger based checks:
-You cast a skill. It will check your stats and cast it.
-You switch gear, it will have to recheck all skills it can apply to and recalculate based on your stats.
-You assign a passive/skill point. It will have to recheck all skills it can apply to and recalculate.
-You get a buff/debuff. Again the same thing.
Anything that changes your stats would require recalculating and re-applying to the skill.

It should also be noted that not snapshotting could lead to cases where you get a debuff and now your minions don’t deal damage because your stats were tanked.

Because games now do a lot more than they ever did before? LE is much more complex mechanically than D2 ever was. It has way more moving parts.
Also D2 was much more limited in what the server did. Almost all the code ran in your PC. The server was mostly for syncs (which is also why you could cheat so much, even online).

As I said, falconer isn’t really a minion. It can’t be targetted. Stats like minion health, resists, armor, etc, can all be ignored.
When the falconer attacks, it’s just like using a skill yourself, so it will calculate again.

1 Like
18

Blockquote Trigger based checks:
-You cast a skill. It will check your stats and cast it.
-You switch gear, it will have to recheck all skills it can apply to and recalculate based on your stats.
-You assign a passive/skill point. It will have to recheck all skills it can apply to and recalculate.
-You get a buff/debuff. Again the same thing.
Anything that changes your stats would require recalculating and re-applying to the skill.

And that would effect performance to an unplayable degree? So much so that it’s better to let the game be worse because of it? If so, then remove cast/attack speed, and all cast on cast abilities because it makes the performance worse. How about we just delete the game from our computers so it performs better?
Just because you don’t care about snapshotting, doesn’t mean other people don’t too. I hear it all the time that players don’t even want to get into a build because of snapshotting.

Is that not a valid reason for changing the way the certain skills operate? It’s not a small or isolated problem either. Snapshotting indirectly effects balance changes (and don’t tell me it doesn’t when we have balance choices such as the new idols; referring to the incompetency of EHG balance), it directly effects build guides, and directly effects any sort of competitive nature of the game.

I will die on this hill. The years of ignored snapshotting issues being necessary for performance is archaic. The idea of “it’s too much work” is ignorant when EHG have had years of development re-think and (even if slowly) implement a solution to an issue that isn’t new.

19

No one ever said it’s better to leave it as is. Just that a solution needs to be found that won’t affect the performance. Which is what the devs are trying to do.
And simply adding checks to every minion/toggleable skill isn’t a solution.

It is. But changing the way skills work entirely will lead to a massive undertaking that will require lots of time and effort.

So maybe they can figure out a different way to address this. For example, you could turn all relevant stat affixes into a global modifier that is added to the minion’s attack and defenses. This way, there is no extra check, you’re just adding variables to the formula.
It would still require some changing of the game, but likely less so than reworking the way skills work entirely.

In all those years they weren’t idle though. They have churned things out all the time.

And you can’t really make something like this be implemented “slowly”. It’s a change. It needs to happen, but you can’t really do it in small bits and pieces.
It would be like fixing your car slowly by taking out one tire or a piece of the engine so you can keep using it.

20

That’s fair enough & when you make your own game & have to deal with the consequences of your choices as every company/person/etc has to. Then you’ll get to have entitled arsehats valued customers come & give you shit for the decisions you made in good faith to get your stuff working & out the door & tell you that you’re a stupid lazy dev & that a fix for it is trivial. It’ll be fun.

1 Like