Vitality -RR% on Nightblade's Night's Chill

Grim Dawn Public Test Discussion
89

The build is good if you enjoy playing it, nothing else matters for the select player experience.

Balancing the game is a grand design decision. You either do it or you don’t. If you do it, a lot of builds will be impacted by the changes, and some builds may even die. Devs try to avoid it as much as possible. PoE has leagues, and some of the issue is mitigated by non-league. GD doesn’t have them, every change is in live version eventually, so it’s more sensitive here.

You absaolutely don’t have to move on. If you have a robust argument against some change that might kill off some builds, hurt diversity more than it’d help, you should use this forum section to appeal.

And home-brewed builds don’t have to be bad. The game has vast building capacity, not all strong builds can be predicted on the design stage. When you go around the intended designed and make smth strong, it’s a good feeling. 90% on my builds are home-brewed, if they were all doomed i wouldn’t try to make them.

5 Likes
90

Among the top builds*

Definitely broken strong and better than any other PB build.

You really can’t tell a good build from a bad build? Well I can explain. Build can be good because of the skills/gear/devotion synergies. Build can be also good because you have made close to optimal choices when choosing gear, affixes, stat distribution, devotion map, devotion bindings etc. etc. etc.

It’s not rocket science. It’s also not some kind of art where everything is subjective and you can just stand there smirking while twisting your French moustache, smoking a cigarrete with a cigarrete holder saying “But why can’t homebrew build be the best?” or something. Once you have enough experience in this game especially in endgame, you can differentiate between a bad build and a good build. More experience you get the better your understanding. Some players spent a lot of hours in this game figuring endgame builds.

It’s really that simple.

2 Likes
91

super easy comparison, imo, would be if you take what’s on appearance a “Logical” decent build, it seems to tick all the right boxes, like phys AAR battlemage
You have the legends, you might have some acceptable/default fitting phys dmg greens, you “accidentally” stumbled on getting a bit of RR be it deathstalker or combust band whatever, devos are not totally Gnomish conversion meme absurd
And yet it just falls totally behind ex Avenger warder, another “similar” physical dmg build with no bonus class RR etc.
You can “stumble” on the same build, you can in your best way try to optimize might as you feel it may fit/be “best”.
And you have a simple 5min build with lunacy survival on one hand, and a 7+min build with “regular”/non spectactular survival on the other

that’s like a stark contrast since they technically isnt’ the same build besides being phys dmg on single RR class. But that’s like in the area where it’s so obvious when you compare the metrics even on the more casual pilot front you usually start to be able to tell that one skill/build support is vastly differently than another

I’m sure as heck not good enough to notice the slight nuance difference between a ±10 or 30sec same build optimization/item+devo change, or if i’m getting fucked over because i had 60% slow res instead of 80, or getting impeded by lack of fumble in build or too much fumble on general enemies (grava ex excluded).
But i also don’t get the sense that’s what we’re talking about in terms of separating “good builds from bad”, but that the experienced players are making references more applicable to the casual player example of the 5-7min ± noticeable survival differences, even if it’s not applied to builds with the same huge/stark gap. Eg a 2+min casual player difference/notice might be a 45sec-1min difference paid attention to on the high end/by experienced pilots. Likewise survival notes might not be as wide apart, but the difference if the build comfortably facetanks 4 bosses vs 3 or 2, or if it handles 90-100% boss combos multi-tank without issue vs just 60%, where the casual might never notice simply always 1v1’ing.
Hell i wasn’t even aware of Grava’s charge disrupt for ages until someone pointed it out as one his annoying features in cruci, i’d assumed it was just stun :smile:

It’s one of the reasons i never try to compare by chunk clear speeds to nofika/roman etc because i’m aware of like my innate drawbacks of not moving on quick enough or dawdling around locating targets.
But if i’m going from a regular approach of feeling sweaty 1v1/not adjusting my boss fight habit of solo fighting, to suddenly no braining 3bosses at the same time, then even i can tell the “bad” build from a better build. Better pilots/more experienced players just need slightly less nuance to tell those differences apart is all.

2 Likes
92

(except for Battlemage where -rr% aura can’t be added).

Why can’t Battlemages get -RR% exactly?

93

Only skill that can have that RR in Arcanist is Flash Freeze and that has to actually freeze to apply the RR. And none of Soldier skills can have that RR at all.

94

lack of debuffs
-x% RR has to be attached to psuedo dot/ticking debuffs, it’s why wind devils and thermite mines basically is just wearing aura of censure, which is a debuff class aura vs ex buff(field command) or attack aura like counter strike
It’s why the only -x% RR modifiers you see on battlemages, or more specifically on arcanist is to OFF; because it is actually doing a debuff on main node

and what Norzan says is also true, Flash Freeze does a freeze res check first, if enemy can’t be frozen the debuff wont be applied at all, and will only last/tick for as long as the enemy is frozen (so high freeze res enemies gets drastic reduced RR duration)

1 Like
95

Im guessing its figuring out where putting that rr on skills and what types of those rrs make sense for soldier or arcanist.

The rrs that would be suited for a mage if it was is physical, bleeding, elemental and aether

96

Yeah, but in saying that, that’s all down to player skill.

I can’t get close to 4 minutes with any of those top 20 builds simply because my brain’s a bit shit. Never mind other player’s ability and skills. So run time is always going to be variable unless the best players are testing them.

Which means those times are not the best guide to how a build will handle in the hands of others.

Hell, personally I’d prefer a 20 run average as the base line, simply because the presence/absence of 1 or more short rooms can radically alter the run time. And if I were more functional I’d be using that myself. But then, I have stats papers under my belt, so I’m going to look at you saying:

And laugh. Because this really is more of a qualitative thing subject to player’s capabilities than hard empirical data. Especially when it comes down to how a build feels to play.

Right, back to phy. DB testing. Brain’s extra dead today, so it’s derp-city on my end. Do think Myth. Chausses of Barbaross can be made to work though…

97

That’s down to how skills are coded in game. From the deeps of my memory, basically if it’s set up without -RR% from the start, Crate can’t add it, only flat -RR.

Which is how PRM came to have flat -RR added to it and why Scion of the Elements only adds flat -RR to Cadence.

98

Is there any other objective way to measure a build’s capabilities, other than time to clear SR/Crucible/etc? I mean, the fact that you (or me for that matter) don’t achieve the times posted doesn’t mean anything: if X build takes 3 min on a “skilled” player but it takes 6 for you, and then Y build takes 6 min for the same “skilled” player, it will probably take 12 min for you.

But if a “skilled” player is testing X build and got 3 min runs, then a patch happens and that same player gets 6 min runs for that same build, it’s obvious that something happened, not much to explain. It’s as close to objective as data can get on a scenario like this, so unless there’s a better model to test builds around (which I haven’t seen anyone propose btw), measuring time it’s just the best course of action, imo.

1 Like
99

Objectively? Well, if we go into the philosophical weeds of empiricism, that question becomes a very tangled bank of mind-fuckery. One that’s made many an undergrad (and post grad) philosophy student make an arse out of themselves on what and how objectivity, fact and knowledge art…

Anyhow, if you want 100% truth, logic systems are the only place for such a thing :stuck_out_tongue: Until you run into Godel Incompleteness and potentially loose your sanity that is.

In this particular case, I’d say we can get to an effective consensus via the best players trying things out in an ad-hoc way using shortest kill times vis Supers/CR/SR as their base line. Which is in all truth “good enough” for most players. Ideally we’d also get the worst times posted for SR, which are more close to your average player’s thing. At least for the section of the community that does SR30-31. Which is a lot less than you’d think.

Getting something better would require multiple runs (60+) by multiple players, collating all that data and doing basic statistical analysis on it. Which given the vast majority of the player base, who aren’t autistic weirdo’s/data nerds and don’t have neither the time, nor inclination for that so it isn’t going to happen.

100

Ehm, what’s your point? Isn’t that what’s being done already? Uh?!?! o.O

I don’t see the value in posting the worst times (also by who? the build poster? I think some comment their different times, some don’t; by other people trying the build? then it’s on those people to post them, and they already have the space to do so). A build poster can’t account for every single person’s different skill, but those people trying the build can account for their own skill and conclude that, if X skilled player got let’s say 4 minutes in their run, then someone trying out the build will get 4 minutes + an amount according to their skill level. Besides, from a balance-wise perspective, I think it’s just way easier to account for all the variables when looking at the high end of anything, which has a ceiling (what the minmax community strive for), than looking at the low end, because it’s just never ending; do we buff a build because Timmy insists on putting only 1 point in Shaman and maxing Savagery? No we don’t.

Again, I don’t know what’s the point here. Yes, we all can look at, for example, Steam’s achievements and see that barely 6.8% of the people have completed the base game on Ultimate (without even accounting for HC, mind you). Afaik, SR30-31 is the agreed upon run to maximize the loot drop per time invested, since higher shards won’t really mean better loot, but this doesn’t mean that builds can’t go higher. I think most minmaxed builds can do higher (and if after minmaxing it’s still underperforming, it can be shared so more people can either help minmaxing further or evaluate an actual problem that might need the devs attention). Again, look my previous point about Timmy’s character building.

I think there’s a bit of a misunderstanding; the fact that builds are balanced around the high end of players/skill/times/etc., doesn’t mean that Crate won’t consider new or lower skill players when balancing stuff; if anything, the leveling process is way easier now than a couple years ago (in a good way I mean), farming gear has also been buffed with MIs always dropping from bosses and better affinity between affixes and their gear, even the recent addition of RR to some of the masteries open up more possibilities for homebrew builds without requiring specific gear.

In the end, imo, it just comes down to knowing your own skill level; for example, I know that if I see some of the people here posting a 4 minute run on a build, I will most likely get around 6:30 to 7:00, maybe adapt some more stuff since I play on HC only, etc. I would end up with a homebrew version that will most likely have higher times than the minmaxed version but it would feel better for my personal playstyle. Should my slower, homebrew version, be taken into account for balancing the game? I don’t think so, since I’m taking the informed decision of, for example, sacrificing some damage for defense or something similar.

1 Like
101

Going back on topic, adding the Vit RR to VoS instead of NC is a great start imo, and maybe Demonslayer could get +1 to NB on full set completition to not mess with other builds using partial Demoslayer, but it definetely needs some love. I also found a bit hard to max out Nether Edge without going for “weird” options like Kravall (weird conversion, even if PB already has local conversion), or Dreadchill gloves / Anubar ring (both meant for cold variant). I don’t think those items should be changed to accomodate for vitality builds, but at least some more options would be very welcomed.

102

You know, I made this whole nuanced post with empirical data explaining the optimization process to make you all look good and then you post subjective personal attacks like this. You really aren’t beating the toxic builders allegations here.

I’ll put things in simpler terms. A build is measured in two categories:

  1. How much offensive DPS can the build output?
  2. How much incoming DPS can the build survive?

When leech is involved, points 1 & 2 blend together as the more DPS you output, the more leech you sustain and thus can withstand higher enemy DPS. Build optimization involves figuring out exactly how much incoming enemy DPS you can handle, and then taking the lines through Crucible waves / SR chunks that gather as many enemies as possible within the incoming enemy DPS standard and defeat them using your offensive output.

If your build is not able to withstand nearly as much DPS, you have to take different lines in order to survive, bunch together smaller groups of enemies and approach battles more cautiously, and each sacrifice you make on that front adds more and more time to a build’s “potential” runtime.

Yes, it’s funny to see a top builder slam into a 5 boss group and die instantly, declare the build mid, and have the rest of us laugh and say “durr-hurr, don’t jump into a 5 boss group then,” but that’s simply how the optimization process goes - find a line the top builds can reach, and then show the data to Zantai and the dev team and let them decide if that’s the standard measure builds should compare to and buff/nerf accordingly.

My particular beef that I outlined in my “Everything in One Button” thread is that when builds reach a certain amount of optimization, any amount of variance stemming from unique build approaches or enemy placement becomes intolerable because what works for one type of build would get another build faceplanted. It’s easy to demand changes to make everything the same so that builds can “compete on an equal footing,” but there’s a reason every game reaches a point where you “optimize the fun out of everything,” because introducing a unique mechanic takes time and effort to reach an optimization threshold so that the two mechanics can be fairly compared, and some people just don’t have the time or patience to do that.

3 Likes
103

It’s almost like something like this was done by a group of experienced players, what was it’s name?.. I forgot… Well something about some garlic or something. :scorv:

I honestly was starting to lose hope that sane people like this exist. Really good and simple explanation too, thank you!

I think the solution I found in this test patch - Spellbreaker belt + Kra’Vall shoulders - is the best. Kra’Vall shoulders have conversion that doesn’t affect anything except some crumbs of Chaos getting converted to Cold on build’s weapon damage procs (usually just pistol’s proc and Bat). The offer insane skill points and very useful Freeze res, plus their affix bias is not as harsh as on some other elemental items so getting decent rare affix is easy (and a lot would work). Spellbreaker misses core damage but again, useful skill points and that Nether Edge Bonus. And then you can take juicy Cursebearer and Signet of the Fallen.

This is where you were absolutely right

This is where you misunderstand or/and misinterpret how we do feedback. It’s like this argument I keep hearing “you want everything top20”. I do not. It doesn’t even make sense if you look at our actual top20 testing: we have dozens of builds in our grimsheet that didn’t make top20 and there are dozens more that were played but not included into the testing at all. Did we ask to buff all those build to top20 status? Did we really ask that all of them should be facetanking 4 bosses or bust? I mean why do you think that our approach is so crude that we can’t understand that some builds have limits and noone expects Fire Blightlord Oppressor to be a Crucible champion or some custom made FoI Paladin to facetank four bosses in SR consistently.

2 Likes
104

:stuck_out_tongue:

Well duh, but you’re not doing the full stats on it, just average + minimum run times. But you guys are doing some good work non-the-less, just not the full crunchy stats I crave. But like I said, good enough for the player base’s needs :3

105

Yeah, I explained why, two simple reasons:

  • Most if not all testers wouldn’t have time to run every build that many times
  • Once you have that much experience you don’t need to test that many different variations that many times to say if they are good or not
1 Like
107

easy objective metrics for if build is good or bad:

can it kill everything?
can it facetank everything?
is it lazy?
can it be played drunk?

:scorv:

4 Likes
108

I theoretical craft a cold PB, seems like it unmatched vs vit PB, waiting for someone can test with toon

1 Like
109

Decided to test a more off-beat Vitality Nightblade version, also (surprise surprise) a Reaper.

It’s a pseudo-spam Bone Harvest build (0.5 seconds cd)

изображение

Tank is great, aoe is also decent but single target damage is meme tier despite all the %damage and rr on paper (and Blade Spirits!). Really not sure if that -tdm% on Bonescythe deserves a little nudge or not, but single target dps with that axe seems to be severely lacking. Here are videos with its performance:

1:39 Ravager kill
1:59 Callagadra kill
5:32 SR 30-31 run

2 Likes